wip

- Use cryptographically secure ID generation with node:crypto
- Implement runtime validation for JSON parsing using Zod
- Optimize circular dependency validation to avoid N+1 file reads

.gemini/skills/docs-writer/SKILL.md

@@ -118,8 +118,6 @@ documentation.
   reflects existing code.
 - **Structure:** Apply "Structure (New Docs)" rules (BLUF, headings, etc.) when 
   adding new sections to existing pages.
-- **Headers**: If you change a header, you must check for links that lead to
-  that header and update them.
 - **Tone:** Ensure the tone is active and engaging. Use "you" and contractions.
 - **Clarity:** Correct awkward wording, spelling, and grammar. Rephrase
   sentences to make them easier for users to understand.
@@ -135,8 +133,7 @@ and that all links are functional.
   technical behavior.
 2.  **Self-review:** Re-read changes for formatting, correctness, and flow.
 3.  **Link check:** Verify all new and existing links leading to or from modified
-    pages. If you changed a header, ensure that any links that lead to it are
-    updated.
+    pages.
 4.  **Format:** Once all changes are complete, ask to execute `npm run format`
     to ensure consistent formatting across the project. If the user confirms,
     execute the command.

.gemini/skills/string-reviewer/SKILL.md

@@ -1,99 +0,0 @@
----
-name: string-reviewer
-description: >
-  Use this skill when asked to review text and user-facing strings within the codebase. It ensures that these strings follow rules on clarity,
-  usefulness, brevity and style.
----
-
-# String Reviewer
-
-## Instructions
-
-Act as a Senior UX Writer. Look for user-facing strings that are too long,
-unclear, or inconsistent. This includes inline text, error messages, and other
-user-facing text.
-
-Do NOT automatically change strings without user approval. You must only suggest
-changes and do not attempt to rewrite them directly unless the user explicitly
-asks you to do so.
-
-## Core voice principles
-
-The system prioritizes deterministic clarity over conversational fluff. We
-provide telemetry, not etiquette, ensuring the user retains absolute agency..
-
-1. **Deterministic clarity:** Distinguish between certain system/service states
-   (Cloud Billing, IAM, the System) and probabilistic AI analysis (Gemini).
-2. **System transparency:** Replace "Loading..." with active technical telemetry
-   (e.g., Tracing stack traces...). Keep status updates under 5 words.
-3. **Front-loaded actionability:** Always use the [Goal] + [Action] pattern.
-   Lead with intent so users can scan left-to-right.
-4. **Agentic error recovery:** Every error must be a pivot point. Pair failures
-   with one-click recovery commands or suggested prompts.
-5. **Contextual humility:** Reserve disclaimers and "be careful" warnings for P0
-   (destructive/irreversible) tasks only. Stop warning-fatigue.
-
-## The writing checklist
-
-Use this checklist to audit UI strings and AI responses.
-
-### Identity and voice
-- **Eliminate the "I":** Remove all first-person pronouns (I, me, my, mine).
-- **Subject attribution:** Refer to the AI as Gemini and the infrastructure as
-  the - system or the CLI.
-- **Active voice:** Ensure the subject (Gemini or the system) is clearly
-  performing the action.
-- **Ownership rule:** Use the system for execution (doing) and Gemini for
-  analysis (thinking)
-
-### Structural scannability
-- **The skip test:** Do the first 3 words describe the user’s intent? If not,
-  rewrite.
-- **Goal-first sequence:** Use the template: [To Accomplish X] + [Do Y].
-- **The 5-word rule:** Keep status updates and loading states under 5 words.
-- **Telemetry over etiquette:** Remove polite filler (Please wait, Thank you,
-  Certainly). Replace with raw data or progress indicators.
-- **Micro-state cycles:** For tasks $> 3$ seconds, cycle through specific
-  sub-states (e.g., Parsing logs... ➔ Identifying patterns...) to show momentum.
-
-
-### Technical accuracy and humility
-- **Verb signal check:** Use deterministic verbs (is, will, must) for system
-  state/infrastructure.
-  - Use probabilistic verbs (suggests, appears, may, identifies) for AI output.
-- **No 100% certainty:** Never attribute absolute certainty to model-generated
-  content.
-- **Precision over fuzziness:** Use technical metrics (latency, tokens, compute) instead of "speed" or "cost."
-- **Instructional warnings:** Every warning must include a specific corrective action (e.g., "Perform a dry-run first" or "Review line 42").
-
-### Agentic error recovery
-- **The one-step rule:** Pair every error message with exactly one immediate
-  path to a fix (command, link, or prompt).
-- **Human-first:** Provide a human-readable explanation before machine error
-  codes (e.g., 404, 500).
--  **Suggested prompts:** Offer specific text for the user to copy/click like
-   “Ask Gemini: 'Explain this port error.'”
-
-### Use consistent terminology
-
-Ensure all terminology aligns with the project [word
-list](./references/word-list.md). 
-
-If a string uses a term marked "do not use" or "use with caution," provide a
-correction based on the preferred terms.
-
-## Ensure consistent style for settings
-
-If `packages/cli/src/config/settingsSchema.ts` is modified, confirm labels and
-descriptions specifically follow the unique [Settings
-guidelines](./references/settings.md).
-
-## Output format
-When suggesting changes, always present your review using the following list
-format. Do not provide suggestions outside of this list..
-
-```
-1. **{Rationale/Principle Violated}**
-  - ❌ "{incorrect phrase}"
-  - ✅ `"{corrected phrase}"`
-```

.gemini/skills/string-reviewer/references/settings.md

@@ -1,28 +0,0 @@
-# Settings
-
-## Noun-First Labeling (Scannability)
-
-Labels must start with the subject of the setting, not the action. This allows
-users to scan for the feature they want to change.
-
-- **Rule:** `[Noun]` `[Attribute/Action]`
-- **Example:** `Show line numbers` becomes simply `Line numbers`
-
-## Positive Boolean Logic (Cognitive Ease)
-
-Eliminate "double negatives." Booleans should represent the presence of a
-feature, not its absence.
-
-- **Rule:** Replace `Disable {feature}` or `Hide {Feature}` with
-  `{Feature} enabled` or simply `{Feature}`.
-- **Example:** Change "Disable auto update" to "Auto update".
-- **Implementation:** Invert the boolean value in your config loader so true
-  always equals `On`
-
-## Verb Stripping (Brevity)
-
-Remove redundant leading verbs like "Enable," "Use," "Display," or "Show" unless
-they are part of a specific technical term.
-
-- **Rule**: If the label works without the verb, remove it
-- **Example**: Change `Enable prompt completion` to `Prompt completion`

.gemini/skills/string-reviewer/references/word-list.md

@@ -1,61 +0,0 @@
-## Terms
-
-### Preferred
-
-- Use **create** when a user is creating or setting up something.
-- Use **allow** instead of **may** to indicate that permission has been granted
-  to perform some action.
-- Use **canceled**, not **cancelled**.
-- Use **configure** to refer to the process of changing the attributes of a
-  feature, even if that includes turning on or off the feature.
-- Use **delete** when the action being performed is destructive.
-- Use **enable** for binary operations that turn a feature or API on. Use "turn
-  on" and "turn off" instead of "enable" and "disable" for other situations.
-- Use **key combination** to refer to pressing multiple keys simultaneously.
-- Use **key sequence** to refer to pressing multiple keys separately in order.
-- Use **modify** to refer to something that has changed vs obtaining the latest
-  version of something.
-- Use **remove** when the action being performed takes an item out of a larger
-  whole, but doesn't destroy the item itself.
-- Use **set up** as a verb. Use **setup** as a noun or adjective.
-- Use **show**. In general, use paired with **hide**.
-- Use **sign in**, **sign out** as a verb. Use **sign-in** or **sign-out** as a
-  noun or adjective.
-- Use **update** when you mean to obtain the latest version of something.
-- Use **want** instead of **like** or **would like**.
-
-#### Don't use
-
-- Don't use **etc.** It's redundant. To convey that a series is incomplete,
-  introduce it with "such as" instead.
-- Don't use **hostname**, use "host name" instead.
-- Don't use **in order to**. It's too formal. "Before you can" is usually better
-  in UI text.
-- Don't use **one or more**. Specify the quantity where possible. Use "at least
-  one" when the quantity is 1+ but you can't be sure of the number. Likewise,
-  use "at least one" when the user must choose a quantity of 1+.
-- Don't use the terms **log in**, **log on**, **login**, **logout** or **log
-  out**.
-- Don't use **like** or **would you like**. Use **want** instead. Better yet,
-  rephrase so that it's not referring to the user's emotional state, but rather
-  what is required.
-
-#### Use with caution
-
-- Avoid using **leverage**, especially as a verb. "Leverage" is considered a
-  buzzword largely devoid of meaning apart from the simpler "use".
-- Avoid using **once** as a synonym for "after". Typically, when "once" is used
-  in this way, it is followed by a verb in the perfect tense.
-- Don't use **e.g.** Use "example", "such as", "like", or "for example". The
-  phrase is always followed by a comma.
-- Don't use **i.e.** unless absolutely essential to make text fit. Use "that is"
-  instead.
-- Use **disable** for binary operations that turn a feature or API off. Use
-  "turn on" and "turn off" instead of "enable" and "disable" for other
-  situations. For UI elements that are not available, use "dimmed" instead of
-  "disabled".
-- Use **please** only when you're asking the user to do something inconvenient,
-  not just following the instructions in a typical flow.
-- Use **really** sparingly in such constructions as "Do you really want to..."
-  Because of the weight it puts on the decision, it should be used to confirm
-  actions that the user is extremely unlikely to make.

.github/CODEOWNERS

@@ -14,9 +14,4 @@
 
 # Docs have a dedicated approver group in addition to maintainers
 /docs/ @google-gemini/gemini-cli-maintainers @google-gemini/gemini-cli-docs
-/README.md @google-gemini/gemini-cli-maintainers @google-gemini/gemini-cli-docs
-
-# Prompt contents, tool definitions, and evals require reviews from prompt approvers
-/packages/core/src/prompts/ @google-gemini/gemini-cli-prompt-approvers
-/packages/core/src/tools/ @google-gemini/gemini-cli-prompt-approvers
-/evals/ @google-gemini/gemini-cli-prompt-approvers
+/README.md @google-gemini/gemini-cli-maintainers @google-gemini/gemini-cli-docs

.github/actions/publish-release/action.yml

@@ -192,13 +192,6 @@ runs:
         INPUTS_CLI_PACKAGE_NAME: '${{ inputs.cli-package-name }}'
         INPUTS_A2A_PACKAGE_NAME: '${{ inputs.a2a-package-name }}'
 
-    - name: '📦 Prepare bundled CLI for npm release'
-      if: "inputs.npm-registry-url != 'https://npm.pkg.github.com/'"
-      working-directory: '${{ inputs.working-directory }}'
-      shell: 'bash'
-      run: |
-        node ${{ github.workspace }}/scripts/prepare-npm-release.js
-
     - name: 'Get CLI Token'
       uses: './.github/actions/npm-auth-token'
       id: 'cli-token'

.github/workflows/chained_e2e.yml

@@ -264,27 +264,6 @@ jobs:
         run: 'npm run build'
         shell: 'pwsh'
 
-      - name: 'Ensure Chrome is available'
-        shell: 'pwsh'
-        run: |
-          $chromePaths = @(
-            "${env:ProgramFiles}\Google\Chrome\Application\chrome.exe",
-            "${env:ProgramFiles(x86)}\Google\Chrome\Application\chrome.exe"
-          )
-          $chromeExists = $chromePaths | Where-Object { Test-Path $_ } | Select-Object -First 1
-          if (-not $chromeExists) {
-            Write-Host 'Chrome not found, installing via Chocolatey...'
-            choco install googlechrome -y --no-progress --ignore-checksums
-          }
-          $installed = $chromePaths | Where-Object { Test-Path $_ } | Select-Object -First 1
-          if ($installed) {
-            Write-Host "Chrome found at: $installed"
-            & $installed --version
-          } else {
-            Write-Error 'Chrome installation failed'
-            exit 1
-          }
-
       - name: 'Run E2E tests'
         env:
           GEMINI_API_KEY: '${{ secrets.GEMINI_API_KEY }}'

.github/workflows/ci.yml

@@ -169,7 +169,7 @@ jobs:
             npm run test:ci --workspace @google/gemini-cli
           else
             # Explicitly list non-cli packages to ensure they are sharded correctly
-            npm run test:ci --workspace @google/gemini-cli-core --workspace @google/gemini-cli-a2a-server --workspace gemini-cli-vscode-ide-companion --workspace @google/gemini-cli-test-utils --if-present -- --coverage.enabled=false
+            npm run test:ci --workspace @google/gemini-cli-core --workspace @google/gemini-cli-a2a-server --workspace gemini-cli-vscode-ide-companion --workspace @google/gemini-cli-test-utils --if-present
             npm run test:scripts
           fi
 

.github/workflows/gemini-automated-issue-triage.yml

@@ -121,7 +121,6 @@ jobs:
               'area/security',
               'area/platform',
               'area/extensions',
-              'area/documentation',
               'area/unknown'
             ];
             const labelNames = labels.map(label => label.name).filter(name => allowedLabels.includes(name));
@@ -256,14 +255,6 @@ jobs:
               "Issues with a specific extension."
               "Feature request for the extension ecosystem."
 
-            area/documentation
-            - Description: Issues related to user-facing documentation and other content on the documentation website.
-            - Example Issues:
-              "A typo in a README file."
-              "DOCS: A command is not working as described in the documentation."
-              "A request for a new documentation page."
-              "Instructions missing for skills feature"
-
             area/unknown
             - Description: Issues that do not clearly fit into any other defined area/ category, or where information is too limited to make a determination. Use this when no other area is appropriate.
 

.github/workflows/gemini-scheduled-issue-triage.yml

@@ -63,7 +63,7 @@ jobs:
 
           echo '🔍 Finding issues missing area labels...'
           NO_AREA_ISSUES="$(gh issue list --repo "${GITHUB_REPOSITORY}" \
-            --search 'is:open is:issue -label:area/core -label:area/agent -label:area/enterprise -label:area/non-interactive -label:area/security -label:area/platform -label:area/extensions -label:area/documentation -label:area/unknown' --limit 100 --json number,title,body)"
+            --search 'is:open is:issue -label:area/core -label:area/agent -label:area/enterprise -label:area/non-interactive -label:area/security -label:area/platform -label:area/extensions -label:area/unknown' --limit 100 --json number,title,body)"
 
           echo '🔍 Finding issues missing kind labels...'
           NO_KIND_ISSUES="$(gh issue list --repo "${GITHUB_REPOSITORY}" \
@@ -204,7 +204,6 @@ jobs:
             Categorization Guidelines (Area):
             area/agent: Core Agent, Tools, Memory, Sub-Agents, Hooks, Agent Quality
             area/core: User Interface, OS Support, Core Functionality
-            area/documentation: End-user and contributor-facing documentation, website-related
             area/enterprise: Telemetry, Policy, Quota / Licensing
             area/extensions: Gemini CLI extensions capability
             area/non-interactive: GitHub Actions, SDK, 3P Integrations, Shell Scripting, Command line automation

.github/workflows/unassign-inactive-assignees.yml

@@ -1,315 +0,0 @@
-name: 'Unassign Inactive Issue Assignees'
-
-# This workflow runs daily and scans every open "help wanted" issue that has
-# one or more assignees.  For each assignee it checks whether they have a
-# non-draft pull request (open and ready for review, or already merged) that
-# is linked to the issue.  Draft PRs are intentionally excluded so that
-# contributors cannot reset the check by opening a no-op PR.  If no
-# qualifying PR is found within 7 days of assignment the assignee is
-# automatically removed and a friendly comment is posted so that other
-# contributors can pick up the work.
-# Maintainers, org members, and collaborators (anyone with write access or
-# above) are always exempted and will never be auto-unassigned.
-
-on:
-  schedule:
-    - cron: '0 9 * * *' # Every day at 09:00 UTC
-  workflow_dispatch:
-    inputs:
-      dry_run:
-        description: 'Run in dry-run mode (no changes will be applied)'
-        required: false
-        default: false
-        type: 'boolean'
-
-concurrency:
-  group: '${{ github.workflow }}'
-  cancel-in-progress: true
-
-defaults:
-  run:
-    shell: 'bash'
-
-jobs:
-  unassign-inactive-assignees:
-    if: "github.repository == 'google-gemini/gemini-cli'"
-    runs-on: 'ubuntu-latest'
-    permissions:
-      issues: 'write'
-
-    steps:
-      - name: 'Generate GitHub App Token'
-        id: 'generate_token'
-        uses: 'actions/create-github-app-token@v2'
-        with:
-          app-id: '${{ secrets.APP_ID }}'
-          private-key: '${{ secrets.PRIVATE_KEY }}'
-
-      - name: 'Unassign inactive assignees'
-        uses: 'actions/github-script@v7'
-        env:
-          DRY_RUN: '${{ inputs.dry_run }}'
-        with:
-          github-token: '${{ steps.generate_token.outputs.token }}'
-          script: |
-            const dryRun = process.env.DRY_RUN === 'true';
-            if (dryRun) {
-              core.info('DRY RUN MODE ENABLED: No changes will be applied.');
-            }
-
-            const owner = context.repo.owner;
-            const repo = context.repo.repo;
-            const GRACE_PERIOD_DAYS = 7;
-            const now = new Date();
-
-            let maintainerLogins = new Set();
-            const teams = ['gemini-cli-maintainers', 'gemini-cli-askmode-approvers', 'gemini-cli-docs'];
-
-            for (const team_slug of teams) {
-              try {
-                const members = await github.paginate(github.rest.teams.listMembersInOrg, {
-                  org: owner,
-                  team_slug,
-                });
-                for (const m of members) maintainerLogins.add(m.login.toLowerCase());
-                core.info(`Fetched ${members.length} members from team ${team_slug}.`);
-              } catch (e) {
-                core.warning(`Could not fetch team ${team_slug}: ${e.message}`);
-              }
-            }
-
-            const isGooglerCache = new Map();
-            const isGoogler = async (login) => {
-              if (isGooglerCache.has(login)) return isGooglerCache.get(login);
-              try {
-                for (const org of ['googlers', 'google']) {
-                  try {
-                    await github.rest.orgs.checkMembershipForUser({ org, username: login });
-                    isGooglerCache.set(login, true);
-                    return true;
-                  } catch (e) {
-                    if (e.status !== 404) throw e;
-                  }
-                }
-              } catch (e) {
-                core.warning(`Could not check org membership for ${login}: ${e.message}`);
-              }
-              isGooglerCache.set(login, false);
-              return false;
-            };
-
-            const permissionCache = new Map();
-            const isPrivilegedUser = async (login) => {
-              if (maintainerLogins.has(login.toLowerCase())) return true;
-
-              if (permissionCache.has(login)) return permissionCache.get(login);
-
-              try {
-                const { data } = await github.rest.repos.getCollaboratorPermissionLevel({
-                  owner,
-                  repo,
-                  username: login,
-                });
-                const privileged = ['admin', 'maintain', 'write', 'triage'].includes(data.permission);
-                permissionCache.set(login, privileged);
-                if (privileged) {
-                  core.info(`  @${login} is a repo collaborator (${data.permission}) — exempt.`);
-                  return true;
-                }
-              } catch (e) {
-                if (e.status !== 404) {
-                  core.warning(`Could not check permission for ${login}: ${e.message}`);
-                }
-              }
-
-              const googler = await isGoogler(login);
-              permissionCache.set(login, googler);
-              return googler;
-            };
-
-            core.info('Fetching open "help wanted" issues with assignees...');
-
-            const issues = await github.paginate(github.rest.issues.listForRepo, {
-              owner,
-              repo,
-              state: 'open',
-              labels: 'help wanted',
-              per_page: 100,
-            });
-
-            const assignedIssues = issues.filter(
-              (issue) => !issue.pull_request && issue.assignees && issue.assignees.length > 0
-            );
-
-            core.info(`Found ${assignedIssues.length} assigned "help wanted" issues.`);
-
-            let totalUnassigned = 0;
-
-              let timelineEvents = [];
-              try {
-                timelineEvents = await github.paginate(github.rest.issues.listEventsForTimeline, {
-                  owner,
-                  repo,
-                  issue_number: issue.number,
-                  per_page: 100,
-                  mediaType: { previews: ['mockingbird'] },
-                });
-              } catch (err) {
-                core.warning(`Could not fetch timeline for issue #${issue.number}: ${err.message}`);
-                continue;
-              }
-
-              const assignedAtMap = new Map();
-
-              for (const event of timelineEvents) {
-                if (event.event === 'assigned' && event.assignee) {
-                  const login = event.assignee.login.toLowerCase();
-                  const at = new Date(event.created_at);
-                  assignedAtMap.set(login, at);
-                } else if (event.event === 'unassigned' && event.assignee) {
-                  assignedAtMap.delete(event.assignee.login.toLowerCase());
-                }
-              }
-
-              const linkedPRAuthorSet = new Set();
-              const seenPRKeys = new Set();
-
-              for (const event of timelineEvents) {
-                if (
-                  event.event !== 'cross-referenced' ||
-                  !event.source ||
-                  event.source.type !== 'pull_request' ||
-                  !event.source.issue ||
-                  !event.source.issue.user ||
-                  !event.source.issue.number ||
-                  !event.source.issue.repository
-                ) continue;
-
-                const prOwner  = event.source.issue.repository.owner.login;
-                const prRepo   = event.source.issue.repository.name;
-                const prNumber = event.source.issue.number;
-                const prAuthor = event.source.issue.user.login.toLowerCase();
-                const prKey    = `${prOwner}/${prRepo}#${prNumber}`;
-
-                if (seenPRKeys.has(prKey)) continue;
-                seenPRKeys.add(prKey);
-
-                try {
-                  const { data: pr } = await github.rest.pulls.get({
-                    owner: prOwner,
-                    repo: prRepo,
-                    pull_number: prNumber,
-                  });
-
-                  const isReady = (pr.state === 'open' && !pr.draft) ||
-                                  (pr.state === 'closed' && pr.merged_at !== null);
-
-                  core.info(
-                    `  PR ${prKey} by @${prAuthor}: ` +
-                    `state=${pr.state}, draft=${pr.draft}, merged=${!!pr.merged_at} → ` +
-                    (isReady ? 'qualifies' : 'does NOT qualify (draft or closed without merge)')
-                  );
-
-                  if (isReady) linkedPRAuthorSet.add(prAuthor);
-                } catch (err) {
-                  core.warning(`Could not fetch PR ${prKey}: ${err.message}`);
-                }
-              }
-
-              const assigneesToRemove = [];
-
-              for (const assignee of issue.assignees) {
-                const login = assignee.login.toLowerCase();
-
-                if (await isPrivilegedUser(assignee.login)) {
-                  core.info(`  @${assignee.login}: privileged user — skipping.`);
-                  continue;
-                }
-
-                const assignedAt = assignedAtMap.get(login);
-
-                if (!assignedAt) {
-                  core.warning(
-                    `No 'assigned' event found for @${login} on issue #${issue.number}; ` +
-                    `falling back to issue creation date (${issue.created_at}).`
-                  );
-                  assignedAtMap.set(login, new Date(issue.created_at));
-                }
-                const resolvedAssignedAt = assignedAtMap.get(login);
-
-                const daysSinceAssignment = (now - resolvedAssignedAt) / (1000 * 60 * 60 * 24);
-
-                core.info(
-                  `  @${login}: assigned ${daysSinceAssignment.toFixed(1)} day(s) ago, ` +
-                  `ready-for-review PR: ${linkedPRAuthorSet.has(login) ? 'yes' : 'no'}`
-                );
-
-                if (daysSinceAssignment < GRACE_PERIOD_DAYS) {
-                  core.info(`    → within grace period, skipping.`);
-                  continue;
-                }
-
-                if (linkedPRAuthorSet.has(login)) {
-                  core.info(`    → ready-for-review PR found, keeping assignment.`);
-                  continue;
-                }
-
-                core.info(`    → no ready-for-review PR after ${GRACE_PERIOD_DAYS} days, will unassign.`);
-                assigneesToRemove.push(assignee.login);
-              }
-
-              if (assigneesToRemove.length === 0) {
-                continue;
-              }
-
-              if (!dryRun) {
-                try {
-                  await github.rest.issues.removeAssignees({
-                    owner,
-                    repo,
-                    issue_number: issue.number,
-                    assignees: assigneesToRemove,
-                  });
-                } catch (err) {
-                  core.warning(
-                    `Failed to unassign ${assigneesToRemove.join(', ')} from issue #${issue.number}: ${err.message}`
-                  );
-                  continue;
-                }
-
-                const mentionList = assigneesToRemove.map((l) => `@${l}`).join(', ');
-                const commentBody =
-                  `👋 ${mentionList} — it has been more than ${GRACE_PERIOD_DAYS} days since ` +
-                  `you were assigned to this issue and we could not find a pull request ` +
-                  `ready for review.\n\n` +
-                  `To keep the backlog moving and ensure issues stay accessible to all ` +
-                  `contributors, we require a PR that is open and ready for review (not a ` +
-                  `draft) within ${GRACE_PERIOD_DAYS} days of assignment.\n\n` +
-                  `We are automatically unassigning you so that other contributors can pick ` +
-                  `this up. If you are still actively working on this, please:\n` +
-                  `1. Re-assign yourself by commenting \`/assign\`.\n` +
-                  `2. Open a PR (not a draft) linked to this issue (e.g. \`Fixes #${issue.number}\`) ` +
-                  `within ${GRACE_PERIOD_DAYS} days so the automation knows real progress is being made.\n\n` +
-                  `Thank you for your contribution — we hope to see a PR from you soon! 🙏`;
-
-                try {
-                  await github.rest.issues.createComment({
-                    owner,
-                    repo,
-                    issue_number: issue.number,
-                    body: commentBody,
-                  });
-                } catch (err) {
-                  core.warning(
-                    `Failed to post comment on issue #${issue.number}: ${err.message}`
-                  );
-                }
-              }
-
-              totalUnassigned += assigneesToRemove.length;
-              core.info(
-                `  ${dryRun ? '[DRY RUN] Would have unassigned' : 'Unassigned'}: ${assigneesToRemove.join(', ')}`
-              );
-            }
-
-            core.info(`\nDone. Total assignees ${dryRun ? 'that would be' : ''} unassigned: ${totalUnassigned}`);

CONTRIBUTING.md

@@ -320,9 +320,11 @@ npm run lint
 
 - Please adhere to the coding style, patterns, and conventions used throughout
   the existing codebase.
-- Consult [GEMINI.md](../GEMINI.md) (typically found in the project root) for
-  specific instructions related to AI-assisted development, including
-  conventions for React, comments, and Git usage.
+- Consult
+  [GEMINI.md](https://github.com/google-gemini/gemini-cli/blob/main/GEMINI.md)
+  (typically found in the project root) for specific instructions related to
+  AI-assisted development, including conventions for React, comments, and Git
+  usage.
 - **Imports:** Pay special attention to import paths. The project uses ESLint to
   enforce restrictions on relative imports between packages.
 

README.md

@@ -77,7 +77,7 @@ See [Releases](./docs/releases.md) for more details.
 
 ### Preview
 
-New preview releases will be published each week at UTC 23:59 on Tuesdays. These
+New preview releases will be published each week at UTC 2359 on Tuesdays. These
 releases will not have been fully vetted and may contain regressions or other
 outstanding issues. Please help us test and install with `preview` tag.
 
@@ -87,7 +87,7 @@ npm install -g @google/gemini-cli@preview
 
 ### Stable
 
-- New stable releases will be published each week at UTC 20:00 on Tuesdays, this
+- New stable releases will be published each week at UTC 2000 on Tuesdays, this
   will be the full promotion of last week's `preview` release + any bug fixes
   and validations. Use `latest` tag.
 
@@ -97,7 +97,7 @@ npm install -g @google/gemini-cli@latest
 
 ### Nightly
 
-- New releases will be published each day at UTC 00:00. This will be all changes
+- New releases will be published each day at UTC 0000. This will be all changes
   from the main branch as represented at time of release. It should be assumed
   there are pending validations and issues. Use `nightly` tag.
 
@@ -301,7 +301,7 @@ gemini
 
 ### Tools & Extensions
 
-- [**Built-in Tools Overview**](./docs/reference/tools.md)
+- [**Built-in Tools Overview**](./docs/tools/index.md)
   - [File System Operations](./docs/tools/file-system.md)
   - [Shell Commands](./docs/tools/shell.md)
   - [Web Fetch & Search](./docs/tools/web-fetch.md)
@@ -323,7 +323,8 @@ gemini
 - [**Enterprise Guide**](./docs/cli/enterprise.md) - Deploy and manage in a
   corporate environment.
 - [**Telemetry & Monitoring**](./docs/cli/telemetry.md) - Usage tracking.
-- [**Tools reference**](./docs/reference/tools.md) - Built-in tools overview.
+- [**Tools API Development**](./docs/reference/tools-api.md) - Create custom
+  tools.
 - [**Local development**](./docs/local-development.md) - Local development
   tooling.
 

docs/adk-replat/outline_short.md

@@ -1,885 +0,0 @@
-# Agent Interface Layer — Type Reference
-
-Framework-agnostic interfaces for gemini-cli to swap between its internal execution loop, ADK, OpenRouter, or any future agent framework via configuration.
-
-**Pattern: `(string & {})`** — Used throughout on discriminator fields like `type`, `reason`, `kind`. This TypeScript trick preserves editor autocomplete for known values while still accepting any string at runtime. Without it, the union collapses to plain `string` and you lose suggestions for known values. Zero runtime cost.
-
----
-
-## Events
-
-### AgentEventCommon
-Base fields on every event flowing between agent and host.
-
-```typescript
-interface AgentEventCommon {
-  /** Unique id for the event. */
-  id: string;
-  /** ISO Timestamp for the time at which the event occurred. */
-  timestamp: string;
-  /** The concrete type of the event. */
-  type: AgentEventType;
-  /** Identifies the subagent thread, omitted for "main thread" events. */
-  threadId?: string;
-  /** Identifies the stream (tool loop execution) this event belongs to. */
-  streamId?: string;
-  /** Optional arbitrary metadata for the event. */
-  _meta?: {
-    /** Source of the event e.g. 'user' | 'ext:{ext_name}/hooks/{hook_name}'. */
-    source?: string;
-    [key: string]: unknown;
-  };
-}
-```
-
-### AgentEventType
-Known event types with open extension.
-
-```typescript
-type AgentEventType =
-  | 'initialize' | 'session_update' | 'message'
-  | 'tool_request' | 'tool_update' | 'tool_response'
-  | 'elicitation_request' | 'elicitation_response'
-  | 'usage' | 'error' | 'stream_end' | 'custom'
-  | (string & {});
-```
-
-### AgentEvents
-Event map using interface + mapped type pattern. Interface (not union) enables declaration merging: any module can add new event types without modifying this file.
-
-```typescript
-interface AgentEvents {
-  /** MUST be the first event emitted in a session. */
-  initialize: Initialize;
-  /** Updates configuration about the current session/agent. */
-  session_update: SessionUpdate;
-  /** Message content provided by user, agent, or developer. */
-  message: Message;
-  /** Tool request issued by the agent. */
-  tool_request: ToolRequest;
-  /** Intermediate progress on long-running tools (ephemeral, not sent to model). */
-  tool_update: ToolUpdate;
-  /** Tool response supplied by the host. */
-  tool_response: ToolResponse;
-  /** Elicitation request to be displayed to the user. */
-  elicitation_request: ElicitationRequest;
-  /** User's response to an elicitation to be returned to the agent. */
-  elicitation_response: ElicitationResponse;
-  /** Reports token usage information. */
-  usage: Usage;
-  /** Report errors. */
-  error: ErrorData;
-  /** Signals why a stream ended. Last event in a stream. */
-  stream_end: StreamEnd;
-  /** Custom events for things not otherwise covered above. */
-  custom: CustomEvent;
-}
-
-/** Mapped type: discriminated union derived from the AgentEvents interface. */
-type AgentEvent<
-  EventType extends keyof AgentEvents = keyof AgentEvents,
-> = AgentEventCommon & AgentEvents[EventType] & { type: EventType };
-```
-
-### Event Payloads
-
-```typescript
-/** Initializes a session by binding it to a specific agent and id.
- *  MUST be the first event emitted. */
-interface Initialize {
-  /** The unique identifier for the session. */
-  sessionId: string;
-  /** The unique location of the workspace (usually an absolute filesystem path). */
-  workspace: string;
-  /** The identifier of the agent being used for this session. */
-  agentId: string;
-  /** The schema declared by the agent that can be used for configuration. */
-  configSchema?: Record<string, unknown>;
-}
-
-/** Updates config such as selected model or session title. */
-interface SessionUpdate {
-  /** If provided, updates the human-friendly title of the current session. */
-  title?: string;
-  /** If provided, updates agent-specific config information. */
-  config?: Record<string, unknown>;
-}
-
-/** Message content. role distinguishes user input, agent output, and
- *  developer-injected context (system prompts, hook output, etc.). */
-interface Message {
-  role: 'user' | 'agent' | 'developer';
-  content: ContentPart[];
-}
-
-/** Tool request issued by the agent. The host decides whether to
- *  execute (after policy check) and returns a ToolResponse. */
-interface ToolRequest {
-  /** A unique identifier for this tool request to be correlated by the response. */
-  requestId: string;
-  /** The name of the tool being requested. */
-  name: string;
-  /** The arguments for the tool. */
-  args: Record<string, unknown>;
-}
-
-/**
- * Used to provide intermediate updates on long-running tools such as subagents
- * or shell commands. ToolUpdates are ephemeral status reporting mechanisms only,
- * they do not affect the final result sent to the model.
- */
-interface ToolUpdate {
-  requestId: string;
-  /** Content to be presented to the user (progress output, etc.). */
-  displayContent?: ContentPart[];
-  /** Content to be sent to the model (if needed). */
-  content?: ContentPart[];
-  /** Structured progress data. */
-  data?: Record<string, unknown>;
-}
-
-/** Tool response supplied by the host after executing a tool call. */
-interface ToolResponse {
-  requestId: string;
-  name: string;
-  /** Content representing the tool call's outcome to be presented to the user. */
-  displayContent?: ContentPart[];
-  /** Multi-part content to be sent to the model. */
-  content?: ContentPart[];
-  /** Structured data to be sent to the model. */
-  data?: Record<string, unknown>;
-  /** When true, the tool call encountered an error that will be sent to the model. */
-  isError?: boolean;
-}
-
-/**
- * Elicitation request to be displayed to the user. Generic mechanism for
- * all cases where an agent needs user input: tool confirmations, auth,
- * model fallback, clarification, etc.
- *
- * `kind` distinguishes the type of interaction so the host/UI can render
- * the right controls and route the response correctly. Michael's original
- * design didn't have `kind` — we added it because gemini-cli has 6 confirmation
- * types, ADK has ToolConfirmation + requestCredential, and Claude has canUseTool.
- * Without `kind`, the host has to parse `_meta` or guess.
- */
-interface ElicitationRequest {
-  /** A unique ID for the elicitation request, correlated in response. */
-  requestId: string;
-  /** What kind of elicitation this is. */
-  kind: ElicitationKind;
-  /**
-   * Whether the elicitation should be displayed as part of the message stream or
-   * as a standalone dialog box.
-   */
-  display?: 'inline' | 'modal' | (string & {});
-  /** The question / content to display to the user. */
-  message: string;
-  /** An optional heading/title for longer-form elicitation requests. */
-  title?: string;
-  /** JSON Schema describing what the response should look like.
-   *  Makes the elicitation self-describing — UIs can render dynamically. */
-  requestedSchema?: JsonSchema;
-  /** Structured context for the UI (tool name, args, diff preview, etc.). */
-  context?: Record<string, unknown>;
-  _meta?: Record<string, unknown>;
-}
-
-/** User's response to an elicitation to be returned to the agent. */
-interface ElicitationResponse {
-  requestId: string;
-  action: 'accept' | 'decline' | 'cancel' | (string & {});
-  /** The user's response data, validated against requestedSchema. */
-  content?: Record<string, unknown>;
-  _meta?: Record<string, unknown>;
-}
-
-/** Reports token usage information.
- *  `cost` added because Claude SDK tracks total_cost_usd — cost visibility
- *  per-turn is essential as agents become production infrastructure. */
-interface Usage {
-  model?: string;
-  inputTokens?: number;
-  outputTokens?: number;
-  cachedTokens?: number;
-  /** Cost in USD for this usage increment. */
-  cost?: { usd?: number };
-}
-
-/** Report errors. */
-interface ErrorData {
-  /** Error status code (e.g. Google RPC codes like 'RESOURCE_EXHAUSTED'). */
-  status: string;
-  /** User-facing message to be displayed. */
-  message: string;
-  /** When true, agent execution is halting because of the error. */
-  fatal: boolean;
-}
-
-/**
- * Signals why a stream ended. Last event in every stream.
- * This is the one signal not covered by other events — the AsyncGenerator
- * completing is ambiguous (done? waiting for input? hit a limit?).
- * stream_end tells you why unambiguously and carries summary data.
- */
-interface StreamEnd {
-  /** Which stream ended. */
-  streamId: string;
-  /** Why this stream ended. */
-  reason: StreamEndReason;
-  /** End-of-stream summary data (cost, usage, turn count, result, etc.). */
-  data?: Record<string, unknown>;
-}
-
-/** CustomEvents are kept in the trajectory but do not have any pre-defined purpose. */
-interface CustomEvent {
-  /** A unique type for this custom event. */
-  kind: string;
-  data?: Record<string, unknown>;
-}
-```
-
-### StreamEndReason
-Why the stream ended. No 'started' (that's `initialize`). No 'handoff' (that's a `transfer_to_agent` tool call). Pausing is implicit (elicitation/tool request events).
-
-```typescript
-type StreamEndReason =
-  | 'completed'   // Agent finished successfully
-  | 'failed'      // Agent encountered an error
-  | 'aborted'     // Agent was cancelled by user/host
-  | 'max_turns'   // Hit turn limit
-  | 'max_budget'  // Hit budget limit
-  | 'max_time'    // Hit time limit
-  | 'refusal'     // Model refused the request
-  | (string & {});
-```
-
-### ElicitationKind
-What type of user interaction the agent needs. Covers gemini-cli's 6 confirmation types, ADK's ToolConfirmation + requestCredential, Claude's canUseTool, and OpenAI's mcp_approval.
-
-```typescript
-type ElicitationKind =
-  | 'tool_confirmation'   // Tool needs user approval before executing
-  | 'model_fallback'      // Model unavailable, user picks alternative
-  | 'auth_required'       // Tool/agent needs credentials
-  | 'clarification'       // Agent needs more info from user
-  | 'user_approval'       // Generic approval (non-tool)
-  | (string & {});
-```
-
----
-
-## Session
-
-### AgentSession
-Runtime handle for an agent execution. Four methods with distinct semantics.
-
-```typescript
-/**
- * Created by SessionFactory.create(). This is the primary abstraction that
- * all agent backends implement — the host interacts with LocalAgent, ADK Runner,
- * OpenRouter, and any future backend through this single interface.
- *
- * Key pattern: stream() is used for BOTH starting AND continuing.
- * The stream ends when the agent needs input (tool result, elicitation).
- * Call stream() again with that input to get the next stream.
- * This matches ADK (call runAsync() again) and Michael (call send() again).
- */
-interface AgentSession {
-  /** Start or continue execution. Returns event stream + controller.
-   *  First call: pass initial input and options to start.
-   *  Subsequent calls: pass tool results, elicitation responses, or
-   *  follow-up messages to continue after the previous stream ended. */
-  stream(data: SessionStreamData, options?: ExecutionOptions): AgentResponse;
-
-  /** Change session config mid-stream. Fire-and-forget — does NOT
-   *  affect the event stream directly. Takes effect immediately. */
-  update(config: SessionUpdateData): void;
-
-  /** Intervene in the running generation. Fire-and-forget.
-   *  Can inject content (redirect the model) or trigger actions
-   *  (move to background, etc.). Broader than just content injection. */
-  steer(data: SteerData): void;
-
-  /** Cancel the current stream or entire execution. */
-  abort(options?: { streamId?: string; reason?: string }): void;
-}
-```
-
-- **`stream()`** — Start or continue execution. Returns event stream + controller. Call again with tool results or elicitation responses to continue after the stream ends.
-- **`update()`** — Change session config mid-stream (permission mode, model, tools). Fire-and-forget.
-- **`steer()`** — Intervene in the running generation (inject content, trigger actions). Fire-and-forget.
-- **`abort()`** — Cancel the current execution. Optional `streamId` to target a specific stream.
-
-### AgentResponse
-Returned by `stream()`. Separates the event stream (read channel) from per-turn control (write channel).
-
-```typescript
-interface AgentResponse {
-  /** The event stream. Yields events until the agent needs input
-   *  (tool result, elicitation) or execution completes. */
-  stream: AsyncGenerator<AgentEvent>;
-  /** Per-turn control. */
-  controller: AgentTurnController;
-}
-
-/** Per-turn control channel.
- *  abort() stops the current turn (model stops generating) but the session
- *  stays alive for the next stream() call. Distinct from session.abort()
- *  which terminates the entire execution. */
-interface AgentTurnController {
-  /** Steer the agent mid-turn. Promise resolves once steering message is sent. */
-  steer?: (message: string | ContentPart[]) => Promise<void>;
-  /** Tell the agent to abort the current generation. */
-  abort: () => void;
-}
-```
-
-### SessionStreamData
-Input to `stream()` for starting or continuing execution. The `kind` field distinguishes new messages from tool results and elicitation responses — this is how the host tells the agent whether it's sending a new user message or resuming after a tool call.
-
-```typescript
-type SessionStreamData =
-  | { kind: 'message'; content: string | ContentPart[] }
-  | { kind: 'tool_result'; requestId: string; result: ToolResultData }
-  | { kind: 'elicitation_response'; requestId: string; response: ElicitationResponseData };
-```
-
-### SessionUpdateData
-Mid-stream config changes via `update()`. These are durable session properties — permission mode, model, tools, title. Separate from `steer()` which affects the running generation.
-
-```typescript
-interface SessionUpdateData {
-  /** Change permission mode mid-session (e.g. switch to auto-accept). */
-  permissionMode?: PermissionMode;
-  /** Switch model mid-session (fallback, user override). */
-  model?: string;
-  /** Enable/disable tools mid-session (MCP connect/disconnect). */
-  tools?: ToolDescriptor[];
-  /** Human-friendly session title. */
-  title?: string;
-  /** Catch-all for agent-specific config. */
-  config?: Record<string, unknown>;
-}
-```
-
-### SteerData
-Mid-stream interventions via `steer()`. Can inject content into the running generation OR trigger actions. Not just messages — actions like Ctrl+B to move to background are steering too.
-
-```typescript
-interface SteerData {
-  /** Target a specific stream (defaults to current). */
-  streamId?: string;
-  /** Content to inject into the running generation. */
-  message?: string | ContentPart[];
-  /** Action to trigger mid-stream. */
-  action?: SteerAction;
-  /** Action-specific data. */
-  data?: Record<string, unknown>;
-}
-
-type SteerAction =
-  | 'background'    // Move execution to background (Ctrl+B)
-  | (string & {});  // extensible
-```
-
-### ExecutionOptions
-Options passed on the first `stream()` call. These are creation-time concerns. Subsequent `stream()` calls only pass `SessionStreamData`. Mid-session changes go through `update()`.
-
-```typescript
-interface ExecutionOptions {
-  /** Tool descriptors available to this execution. */
-  tools?: ToolDescriptor[];
-  /** Model override. */
-  model?: string;
-  /** Whether host wants to execute tools (agent pauses on tool calls). */
-  hostToolExecution?: boolean;
-  /** Streaming mode. */
-  streaming?: boolean;
-  /** Permission mode for this execution. */
-  permissionMode?: PermissionMode;
-  /** Session to resume (opaque ID or full snapshot). */
-  sessionRef?: string | SessionSnapshot;
-  /** Branch from sessionRef instead of continuing in-place. */
-  forkSession?: boolean;
-  /** Run without persisting session state. */
-  ephemeral?: boolean;
-  /** Telemetry context for distributed tracing. */
-  traceContext?: TraceContext;
-  /** Abort signal for cancellation propagation. */
-  signal?: AbortSignal;
-}
-
-/** Permission modes covering gemini-cli (default/autoEdit/yolo/plan)
- *  and Claude SDK (default/acceptEdits/plan/dontAsk/bypassPermissions). */
-type PermissionMode =
-  | 'default'       // Normal — ask user for non-read-only tools
-  | 'auto_edit'     // Auto-approve file edits
-  | 'autonomous'    // Auto-approve all
-  | 'plan'          // Read-only, no mutations
-  | (string & {});
-```
-
----
-
-## Agent Descriptor
-
-### AgentDescriptor
-Static, serializable declaration of an agent's identity and capabilities. Consumed by registry, factory, policy, and UI without running the agent. This is our addition — Michael's doc defines how agents communicate (events, session) but not how they declare themselves.
-
-```typescript
-interface AgentDescriptor {
-  /** Unique identifier. Must be a valid identifier (no spaces, special chars). */
-  name: string;
-  /** Human-friendly display name. */
-  displayName?: string;
-  /** One-line description for model routing and tool descriptions. */
-  description: string;
-  /** Executor type — factory key for creating the AgentSession. */
-  executor: ExecutorType;
-  /** Input schema (JSON Schema). What the agent accepts. */
-  inputSchema: JsonSchema;
-  /** Output schema. What the agent produces. */
-  outputSchema?: JsonSchema;
-  /** Declared capabilities (open string array, not fixed booleans). */
-  capabilities?: AgentCapabilities;
-  /** Tools the agent brings with it (not inherited from parent). */
-  ownTools?: ToolDescriptor[];
-  /** Tools the agent requires from the host. */
-  requiredTools?: string[];
-  /** Sub-agent descriptors (for composite agents). */
-  subAgents?: AgentDescriptor[];
-  /** Execution constraints. */
-  constraints?: AgentConstraints;
-  /** Metadata for discovery, trust, versioning. */
-  metadata?: AgentMetadata;
-}
-
-/** Factory key for creating the right AgentSession implementation. */
-type ExecutorType =
-  | 'local'        // gemini-cli's built-in executor
-  | 'adk'          // ADK-TS Runner
-  | 'remote'       // A2A protocol / remote agent
-  | 'openrouter'   // OpenRouter model-as-agent
-  | 'browser'      // Browser automation agent
-  | (string & {});
-
-/** Open capability declaration. Array instead of Set for JSON serializability.
- *  Common conventions: 'elicitation', 'streaming', 'cancellation',
- *  'host_tool_execution', 'mid_stream_update'. */
-type AgentCapabilities = string[];
-
-interface AgentConstraints {
-  maxTurns?: number;
-  maxTimeMinutes?: number;
-  maxLlmCalls?: number;
-  /** Maximum spend in USD. From Claude SDK's maxBudgetUsd. */
-  maxBudgetUsd?: number;
-}
-
-interface AgentMetadata {
-  /** File path where agent was discovered. */
-  filePath?: string;
-  /** Content hash for trust verification. */
-  hash?: string;
-  /** Whether agent is experimental. */
-  experimental?: boolean;
-  /** Semver version of the agent. */
-  version?: string;
-  /** Interface version this agent targets. */
-  interfaceVersion?: string;
-}
-```
-
----
-
-## Tools
-
-### ToolDescriptor
-How a tool declares itself to agents and the host. Michael's doc has ToolRequest/ToolResponse but no type for declaring a tool — the host needs a standard shape when provisioning tools to an agent.
-
-```typescript
-interface ToolDescriptor {
-  name: string;
-  displayName?: string;
-  description: string;
-  /** JSON Schema for parameters. */
-  parametersSchema: JsonSchema;
-  /** Open key-value hints for policy and UI (readOnly, destructive, etc.). */
-  annotations?: ToolAnnotations;
-}
-
-interface ToolAnnotations {
-  [key: string]: unknown;
-}
-```
-
-### ToolCallRequest / ToolResultData
-Tool call from agent to host, and result back.
-
-```typescript
-/** Tool call request from agent to host. */
-interface ToolCallRequest {
-  /** Correlation ID for matching request to response. */
-  requestId: string;
-  /** Tool name. */
-  name: string;
-  /** Tool arguments. */
-  args: Record<string, unknown>;
-}
-
-/** Tool call result from host to agent. */
-interface ToolResultData {
-  /** Content for the model (what goes into conversation history). */
-  llmContent: ContentPart[];
-  /** Content for the user (what gets displayed). */
-  displayContent?: ContentPart[];
-  /** Whether the tool encountered an error. */
-  isError?: boolean;
-  /** Structured error type. */
-  errorType?: string;
-  /** Additional tool calls triggered by this result. */
-  tailCalls?: ToolCallRequest[];
-}
-```
-
-### ToolProvider
-Host-side tool execution interface.
-
-```typescript
-interface ToolProvider {
-  /** Get all available tool descriptors for an agent. */
-  getToolsForAgent(agentName: string): ToolDescriptor[];
-  /** Execute a tool call (after policy check). */
-  execute(request: ToolCallRequest, context: ToolExecutionContext): Promise<ToolResultData>;
-  /** Check if a tool call is allowed (policy + confirmation). */
-  authorize(request: ToolCallRequest): Promise<PolicyDecision>;
-}
-```
-
----
-
-## Factory
-
-### SessionFactory
-Registry that maps executor types to factory functions that create `AgentSession` instances. Replaces the hard-coded if/else chain in `SubagentToolWrapper.createInvocation()`.
-
-```typescript
-interface SessionFactory {
-  /** Register a factory for an executor type. */
-  register(executorType: string, creator: SessionCreator): void;
-  /** Create a session for a descriptor. The returned session is NOT yet
-   *  running — call session.stream() to start execution. */
-  create(descriptor: AgentDescriptor, context: HostContext): AgentSession;
-  /** Check if an executor type is registered. */
-  has(executorType: string): boolean;
-}
-
-type SessionCreator = (descriptor: AgentDescriptor, context: HostContext) => AgentSession;
-```
-
-### HostContext
-Services the host provides to executors.
-
-```typescript
-interface HostContext {
-  /** Policy engine for tool call authorization. */
-  policyEngine: PolicyEvaluator;
-  /** Hook system for lifecycle interception. */
-  hookSystem: LifecycleInterceptor;
-  /** Elicitation channel for user interactions. */
-  elicitation: ElicitationChannel;
-  /** Tool registry for host-provided tools. */
-  toolRegistry: ToolProvider;
-  /** Model configuration service. */
-  modelConfig: ModelConfigProvider;
-  /** Session service for persistence. */
-  sessionService: SessionProvider;
-  /** Abort signal from parent. */
-  signal?: AbortSignal;
-  /** Telemetry context. */
-  traceContext?: TraceContext;
-}
-```
-
----
-
-## Policy
-
-### PolicyEvaluator
-Host-enforced authorization. The agent has NO say in policy — it is entirely host-enforced. Generic `action` string (not `toolName`) so it works for tools, agents, file access, etc.
-
-```typescript
-interface PolicyEvaluator {
-  /** Evaluate an action against policy. */
-  evaluate(request: PolicyRequest): Promise<PolicyDecision>;
-  /** Get actions statically denied (for pre-filtering without args). */
-  getExcluded(scope?: string): string[];
-}
-
-interface PolicyRequest {
-  /** What is being evaluated (tool name, agent name, resource path, etc.). */
-  action: string;
-  /** Arguments/context for the action. */
-  args?: Record<string, unknown>;
-  /** Who is requesting (agent name, user ID, etc.). */
-  principal?: string;
-  /** Nesting path for hierarchical policy inheritance. */
-  principalPath?: string[];
-  /** Additional context (MCP server name, approval mode, etc.). */
-  context?: Record<string, unknown>;
-}
-
-/** Maps to ADK SecurityPlugin's PolicyOutcome (ALLOW/DENY/CONFIRM)
- *  and gemini-cli's TOML policy decisions. */
-type PolicyOutcome =
-  | 'allow'     // Proceed
-  | 'deny'      // Block
-  | 'ask_user'  // Prompt user for confirmation
-  | (string & {});
-
-type PolicyDecision =
-  | { outcome: 'allow'; message?: string; _meta?: Record<string, unknown> }
-  | { outcome: 'deny'; reason: string; _meta?: Record<string, unknown> }
-  | { outcome: 'ask_user'; hint?: string; _meta?: Record<string, unknown> }
-  | { outcome: PolicyOutcome; _meta?: Record<string, unknown>; [key: string]: unknown };
-```
-
----
-
-## Hooks
-
-### LifecycleInterceptor
-How executors notify the host about lifecycle events so hooks fire regardless of backend. Generic `fire()` with open hook point string rather than separate typed methods — new hook points can be added without changing the interface.
-
-```typescript
-interface LifecycleInterceptor {
-  /** Fire a lifecycle hook by name. */
-  fire(hookPoint: HookPoint, payload: Record<string, unknown>): Promise<HookResult>;
-  /** Check which hook points the host supports. */
-  supportedHookPoints(): HookPoint[];
-}
-
-/** Known hook points from gemini-cli (11) + ADK plugins (12). */
-type HookPoint =
-  | 'before_agent' | 'after_agent'
-  | 'before_model' | 'after_model'
-  | 'before_tool'  | 'after_tool'
-  | 'on_event' | 'on_user_message'
-  | 'before_run' | 'after_run'
-  | 'on_model_error' | 'on_tool_error'
-  | 'before_tool_selection' | 'notification'
-  | 'session_start' | 'session_end' | 'pre_compress'
-  | (string & {});
-
-type HookResult = {
-  /** Should execution continue? */
-  action: 'proceed' | 'block';
-  /** Why blocked (if action is 'block'). */
-  reason?: string;
-  /** Modifications to apply (hook-point-specific). */
-  modifications?: Record<string, unknown>;
-  /** Message for the user (not sent to the model). */
-  systemMessage?: string;
-  _meta?: Record<string, unknown>;
-};
-```
-
----
-
-## Content
-
-### ContentPart
-Provider-agnostic content model used throughout. Aligns with Michael's ContentPart types. `refusal` added based on both Claude SDK (`stop_reason: "refusal"`) and OpenAI (`response.refusal.delta/done`) — model refusals are operationally important. `code`/`code_result` added for code execution agents.
-
-```typescript
-type ContentPart = (
-  /** Represents text. */
-  | { type: 'text'; text: string }
-  /** Represents model thinking output. */
-  | { type: 'thought'; thought: string; thoughtSignature?: string }
-  /** Represents rich media (image/video/pdf/etc) included inline. */
-  | { type: 'media'; mimeType: string; data?: string; uri?: string }
-  /** Represents an inline reference to a resource, e.g. @-mention of a file. */
-  | { type: 'reference'; text: string; uri?: string; mimeType?: string; data?: string }
-  /** Represents a model refusal (safety/policy trigger). */
-  | { type: 'refusal'; text: string }
-  /** Code block. */
-  | { type: 'code'; language: string; code: string }
-  /** Code execution result. */
-  | { type: 'code_result'; output: string; exitCode?: number }
-  /** Function call (for model history serialization). */
-  | { type: 'function_call'; name: string; args: Record<string, unknown>; id: string }
-  /** Function response (for model history serialization). */
-  | { type: 'function_response'; name: string; response: unknown; id: string }
-) & {
-  /** Per-part metadata. Aligns with Michael's ContentPart._meta design. */
-  _meta?: Record<string, unknown>;
-};
-```
-
----
-
-## Model
-
-### ModelInterface
-Provider-agnostic LLM access. Owned by the executor, not the host. No `@google/genai` types — adapters translate at the boundary.
-
-```typescript
-interface ModelInterface {
-  /** Generate content (non-streaming). */
-  generate(request: ModelRequest): Promise<ModelResponse>;
-  /** Generate content (streaming). */
-  generateStream(request: ModelRequest): AsyncGenerator<ModelResponse>;
-}
-
-interface ModelRequest {
-  /** System instruction. */
-  systemInstruction?: ContentPart[];
-  /** Conversation history. */
-  messages: ModelMessage[];
-  /** Available tools. */
-  tools?: ToolDescriptor[];
-  /** Tool calling mode. */
-  toolMode?: 'auto' | 'any' | 'none';
-  /** Generation config. */
-  config?: ModelGenerationConfig;
-}
-
-interface ModelMessage {
-  role: 'user' | 'model' | 'system';
-  content: ContentPart[];
-}
-
-interface ModelResponse {
-  content: ContentPart[];
-  toolCalls?: ToolCallRequest[];
-  usage?: { inputTokens?: number; outputTokens?: number; cachedTokens?: number };
-  /** Raw provider-specific response (escape hatch). */
-  _raw?: unknown;
-}
-
-interface ModelGenerationConfig {
-  temperature?: number;
-  topP?: number;
-  topK?: number;
-  maxOutputTokens?: number;
-  stopSequences?: string[];
-  thinkingConfig?: { thinkingBudget?: number; includeThoughts?: boolean };
-}
-```
-
----
-
-## Persistence
-
-### SessionProvider
-How sessions are persisted and resumed. Out of scope for the initial interface (we implement it, just don't standardize the interface yet) but included here for completeness.
-
-```typescript
-interface SessionProvider {
-  /** Create a new session. */
-  create(agentName: string, metadata?: Record<string, unknown>): Promise<SessionHandle>;
-  /** Load an existing session. */
-  load(sessionId: string): Promise<SessionHandle | undefined>;
-  /** List sessions for an agent. */
-  list(agentName?: string): Promise<SessionSummary[]>;
-  /** Delete a session. */
-  delete(sessionId: string): Promise<void>;
-}
-
-interface SessionHandle {
-  id: string;
-  agentName: string;
-  /** Replay events to reconstruct state (trajectory concept). */
-  events: AgentEvent[];
-  /** Key-value state. */
-  state: Record<string, unknown>;
-  /** Append an event to the session. */
-  append(event: AgentEvent): Promise<void>;
-  /** Get a snapshot for serialization. */
-  snapshot(): SessionSnapshot;
-  lastUpdateTime: number;
-}
-
-interface SessionSnapshot {
-  id: string;
-  agentName: string;
-  events: AgentEvent[];
-  state: Record<string, unknown>;
-}
-
-interface SessionSummary {
-  id: string;
-  agentName: string;
-  title?: string;
-  lastUpdateTime: number;
-  eventCount: number;
-}
-```
-
----
-
-## Composition
-
-### CompositionConfig
-How agents compose into hierarchies, parallel groups, sequential chains, and loops. Maps to ADK's LoopAgent, ParallelAgent, SequentialAgent.
-
-```typescript
-type CompositionPattern =
-  | 'hierarchical'  // Parent calls children as tools
-  | 'sequential'    // Children run in order
-  | 'parallel'      // Children run concurrently
-  | 'loop'          // Children repeat until exit
-  | 'transfer'      // Peer-to-peer handoff via transfer_to_agent tool
-  | (string & {});
-
-interface CompositionConfig {
-  pattern: CompositionPattern;
-  children: AgentDescriptor[];
-  /** For loops: max iterations. */
-  maxIterations?: number;
-  /** For transfers: which agents can be transferred to. */
-  transferTargets?: string[];
-  /** For hierarchical: recursion depth limit. */
-  maxDepth?: number;
-}
-```
-
----
-
-## Telemetry
-
-### TelemetryBridge
-Observability across agent boundaries. Defined in terms of OpenTelemetry concepts but doesn't require OTel as a dependency.
-
-```typescript
-interface TraceContext {
-  /** OpenTelemetry trace ID. */
-  traceId: string;
-  /** Parent span ID. */
-  parentSpanId?: string;
-  /** Baggage items. */
-  baggage?: Record<string, string>;
-}
-
-interface TelemetryBridge {
-  /** Start a span for an agent execution. */
-  startAgentSpan(agentName: string, context?: TraceContext): TelemetrySpan;
-  /** Record a tool call within a span. */
-  recordToolCall(span: TelemetrySpan, toolName: string, args: Record<string, unknown>): void;
-  /** Record a model call. */
-  recordModelCall(span: TelemetrySpan, model: string, usage: Usage): void;
-  /** Record an error. */
-  recordError(span: TelemetrySpan, error: Error): void;
-}
-
-interface TelemetrySpan {
-  /** Create a child span (for sub-agents). */
-  createChild(name: string): TelemetrySpan;
-  /** End the span. */
-  end(status?: 'ok' | 'error'): void;
-  /** Get context for propagation to child executors. */
-  getContext(): TraceContext;
-}
-```

docs/changelogs/latest.md

@@ -1,6 +1,6 @@
-# Latest stable release: v0.32.1
+# Latest stable release: v0.32.0
 
-Released: March 4, 2026
+Released: March 03, 2026
 
 For most users, our latest stable release is the recommended release. Install
 the latest stable version with:
@@ -29,9 +29,6 @@ npm install -g @google/gemini-cli
 
 ## What's Changed
 
-- fix(patch): cherry-pick 0659ad1 to release/v0.32.0-pr-21042 to patch version
-  v0.32.0 and create version 0.32.1 by @gemini-cli-robot in
-  [#21048](https://github.com/google-gemini/gemini-cli/pull/21048)
 - feat(plan): add integration tests for plan mode by @Adib234 in
   [#20214](https://github.com/google-gemini/gemini-cli/pull/20214)
 - fix(acp): update auth handshake to spec by @skeshive in
@@ -205,4 +202,4 @@ npm install -g @google/gemini-cli
   [#19781](https://github.com/google-gemini/gemini-cli/pull/19781)
 
 **Full Changelog**:
-https://github.com/google-gemini/gemini-cli/compare/v0.31.0...v0.32.1
+https://github.com/google-gemini/gemini-cli/compare/v0.31.0...v0.32.0

docs/changelogs/preview.md

@@ -1,6 +1,6 @@
-# Preview release: v0.33.0-preview.4
+# Preview release: v0.33.0-preview.1
 
-Released: March 06, 2026
+Released: March 04, 2026
 
 Our preview release includes the latest, new, and experimental features. This
 release may not be as stable as our [latest weekly release](latest.md).
@@ -29,173 +29,159 @@ npm install -g @google/gemini-cli@preview
 
 ## What's Changed
 
-- fix(patch): cherry-pick 7ec477d to release/v0.33.0-preview.3-pr-21305 to patch
-  version v0.33.0-preview.3 and create version 0.33.0-preview.4 by
-  @gemini-cli-robot in
-  [#21349](https://github.com/google-gemini/gemini-cli/pull/21349)
-- fix(patch): cherry-pick 0135b03 to release/v0.33.0-preview.2-pr-21171
-  [CONFLICTS] by @gemini-cli-robot in
-  [#21336](https://github.com/google-gemini/gemini-cli/pull/21336)
-- fix(patch): cherry-pick 173376b to release/v0.33.0-preview.1-pr-21157 to patch
-  version v0.33.0-preview.1 and create version 0.33.0-preview.2 by
-  @gemini-cli-robot in
-  [#21300](https://github.com/google-gemini/gemini-cli/pull/21300)
-- fix(patch): cherry-pick 173376b to release/v0.33.0-preview.1-pr-21157 to patch
-  version v0.33.0-preview.1 and create version 0.33.0-preview.2 by
-  @gemini-cli-robot in
-  [#21300](https://github.com/google-gemini/gemini-cli/pull/21300)
 - fix(patch): cherry-pick 0659ad1 to release/v0.33.0-preview.0-pr-21042 to patch
   version v0.33.0-preview.0 and create version 0.33.0-preview.1 by
   @gemini-cli-robot in
   [#21047](https://github.com/google-gemini/gemini-cli/pull/21047)
-- Docs: Update model docs to remove Preview Features. by @jkcinouye in
+
+* Docs: Update model docs to remove Preview Features. by @jkcinouye in
   [#20084](https://github.com/google-gemini/gemini-cli/pull/20084)
-- docs: fix typo in installation documentation by @AdityaSharma-Git3207 in
+* docs: fix typo in installation documentation by @AdityaSharma-Git3207 in
   [#20153](https://github.com/google-gemini/gemini-cli/pull/20153)
-- docs: add Windows PowerShell equivalents for environments and scripting by
+* docs: add Windows PowerShell equivalents for environments and scripting by
   @scidomino in [#20333](https://github.com/google-gemini/gemini-cli/pull/20333)
-- fix(core): parse raw ASCII buffer strings in Gaxios errors by @sehoon38 in
+* fix(core): parse raw ASCII buffer strings in Gaxios errors by @sehoon38 in
   [#20626](https://github.com/google-gemini/gemini-cli/pull/20626)
-- chore(release): bump version to 0.33.0-nightly.20260227.ba149afa0 by @galz10
+* chore(release): bump version to 0.33.0-nightly.20260227.ba149afa0 by @galz10
   in [#20637](https://github.com/google-gemini/gemini-cli/pull/20637)
-- fix(github): use robot PAT for automated PRs to pass CLA check by @galz10 in
+* fix(github): use robot PAT for automated PRs to pass CLA check by @galz10 in
   [#20641](https://github.com/google-gemini/gemini-cli/pull/20641)
-- chore/release: bump version to 0.33.0-nightly.20260228.1ca5c05d0 by
+* chore/release: bump version to 0.33.0-nightly.20260228.1ca5c05d0 by
   @gemini-cli-robot in
   [#20644](https://github.com/google-gemini/gemini-cli/pull/20644)
-- Changelog for v0.31.0 by @gemini-cli-robot in
+* Changelog for v0.31.0 by @gemini-cli-robot in
   [#20634](https://github.com/google-gemini/gemini-cli/pull/20634)
-- fix: use full paths for ACP diff payloads by @JagjeevanAK in
+* fix: use full paths for ACP diff payloads by @JagjeevanAK in
   [#19539](https://github.com/google-gemini/gemini-cli/pull/19539)
-- Changelog for v0.32.0-preview.0 by @gemini-cli-robot in
+* Changelog for v0.32.0-preview.0 by @gemini-cli-robot in
   [#20627](https://github.com/google-gemini/gemini-cli/pull/20627)
-- fix: acp/zed race condition between MCP initialisation and prompt by
+* fix: acp/zed race condition between MCP initialisation and prompt by
   @kartikangiras in
   [#20205](https://github.com/google-gemini/gemini-cli/pull/20205)
-- fix(cli): reset themeManager between tests to ensure isolation by
+* fix(cli): reset themeManager between tests to ensure isolation by
   @NTaylorMullen in
   [#20598](https://github.com/google-gemini/gemini-cli/pull/20598)
-- refactor(core): Extract tool parameter names as constants by @SandyTao520 in
+* refactor(core): Extract tool parameter names as constants by @SandyTao520 in
   [#20460](https://github.com/google-gemini/gemini-cli/pull/20460)
-- fix(cli): resolve autoThemeSwitching when background hasn't changed but theme
+* fix(cli): resolve autoThemeSwitching when background hasn't changed but theme
   mismatches by @sehoon38 in
   [#20706](https://github.com/google-gemini/gemini-cli/pull/20706)
-- feat(skills): add github-issue-creator skill by @sehoon38 in
+* feat(skills): add github-issue-creator skill by @sehoon38 in
   [#20709](https://github.com/google-gemini/gemini-cli/pull/20709)
-- fix(cli): allow sub-agent confirmation requests in UI while preventing
+* fix(cli): allow sub-agent confirmation requests in UI while preventing
   background flicker by @abhipatel12 in
   [#20722](https://github.com/google-gemini/gemini-cli/pull/20722)
-- Merge User and Agent Card Descriptions #20849 by @adamfweidman in
+* Merge User and Agent Card Descriptions #20849 by @adamfweidman in
   [#20850](https://github.com/google-gemini/gemini-cli/pull/20850)
-- fix(core): reduce LLM-based loop detection false positives by @SandyTao520 in
+* fix(core): reduce LLM-based loop detection false positives by @SandyTao520 in
   [#20701](https://github.com/google-gemini/gemini-cli/pull/20701)
-- fix(plan): deflake plan mode integration tests by @Adib234 in
+* fix(plan): deflake plan mode integration tests by @Adib234 in
   [#20477](https://github.com/google-gemini/gemini-cli/pull/20477)
-- Add /unassign support by @scidomino in
+* Add /unassign support by @scidomino in
   [#20864](https://github.com/google-gemini/gemini-cli/pull/20864)
-- feat(core): implement HTTP authentication support for A2A remote agents by
+* feat(core): implement HTTP authentication support for A2A remote agents by
   @SandyTao520 in
   [#20510](https://github.com/google-gemini/gemini-cli/pull/20510)
-- feat(core): centralize read_file limits and update gemini-3 description by
+* feat(core): centralize read_file limits and update gemini-3 description by
   @aishaneeshah in
   [#20619](https://github.com/google-gemini/gemini-cli/pull/20619)
-- Do not block CI on evals by @gundermanc in
+* Do not block CI on evals by @gundermanc in
   [#20870](https://github.com/google-gemini/gemini-cli/pull/20870)
-- document node limitation for shift+tab by @scidomino in
+* document node limitation for shift+tab by @scidomino in
   [#20877](https://github.com/google-gemini/gemini-cli/pull/20877)
-- Add install as an option when extension is selected. by @DavidAPierce in
+* Add install as an option when extension is selected. by @DavidAPierce in
   [#20358](https://github.com/google-gemini/gemini-cli/pull/20358)
-- Update CODEOWNERS for README.md reviewers by @g-samroberts in
+* Update CODEOWNERS for README.md reviewers by @g-samroberts in
   [#20860](https://github.com/google-gemini/gemini-cli/pull/20860)
-- feat(core): truncate large MCP tool output by @SandyTao520 in
+* feat(core): truncate large MCP tool output by @SandyTao520 in
   [#19365](https://github.com/google-gemini/gemini-cli/pull/19365)
-- Subagent activity UX. by @gundermanc in
+* Subagent activity UX. by @gundermanc in
   [#17570](https://github.com/google-gemini/gemini-cli/pull/17570)
-- style(cli) : Dialog pattern for /hooks Command by @AbdulTawabJuly in
+* style(cli) : Dialog pattern for /hooks Command by @AbdulTawabJuly in
   [#17930](https://github.com/google-gemini/gemini-cli/pull/17930)
-- feat: redesign header to be compact with ASCII icon by @keithguerin in
+* feat: redesign header to be compact with ASCII icon by @keithguerin in
   [#18713](https://github.com/google-gemini/gemini-cli/pull/18713)
-- fix(core): ensure subagents use qualified MCP tool names by @abhipatel12 in
+* fix(core): ensure subagents use qualified MCP tool names by @abhipatel12 in
   [#20801](https://github.com/google-gemini/gemini-cli/pull/20801)
-- feat(core): support authenticated A2A agent card discovery by @SandyTao520 in
+* feat(core): support authenticated A2A agent card discovery by @SandyTao520 in
   [#20622](https://github.com/google-gemini/gemini-cli/pull/20622)
-- refactor(cli): fully remove React anti patterns, improve type safety and fix
+* refactor(cli): fully remove React anti patterns, improve type safety and fix
   UX oversights in SettingsDialog.tsx by @psinha40898 in
   [#18963](https://github.com/google-gemini/gemini-cli/pull/18963)
-- Adding MCPOAuthProvider implementing the MCPSDK OAuthClientProvider by
+* Adding MCPOAuthProvider implementing the MCPSDK OAuthClientProvider by
   @Nayana-Parameswarappa in
   [#20121](https://github.com/google-gemini/gemini-cli/pull/20121)
-- feat(core): add tool name validation in TOML policy files by @allenhutchison
+* feat(core): add tool name validation in TOML policy files by @allenhutchison
   in [#19281](https://github.com/google-gemini/gemini-cli/pull/19281)
-- docs: fix broken markdown links in main README.md by @Hamdanbinhashim in
+* docs: fix broken markdown links in main README.md by @Hamdanbinhashim in
   [#20300](https://github.com/google-gemini/gemini-cli/pull/20300)
-- refactor(core): replace manual syncPlanModeTools with declarative policy rules
+* refactor(core): replace manual syncPlanModeTools with declarative policy rules
   by @jerop in [#20596](https://github.com/google-gemini/gemini-cli/pull/20596)
-- fix(core): increase default headers timeout to 5 minutes by @gundermanc in
+* fix(core): increase default headers timeout to 5 minutes by @gundermanc in
   [#20890](https://github.com/google-gemini/gemini-cli/pull/20890)
-- feat(admin): enable 30 day default retention for chat history & remove warning
+* feat(admin): enable 30 day default retention for chat history & remove warning
   by @skeshive in
   [#20853](https://github.com/google-gemini/gemini-cli/pull/20853)
-- feat(plan): support annotating plans with feedback for iteration by @Adib234
+* feat(plan): support annotating plans with feedback for iteration by @Adib234
   in [#20876](https://github.com/google-gemini/gemini-cli/pull/20876)
-- Add some dos and don'ts to behavioral evals README. by @gundermanc in
+* Add some dos and don'ts to behavioral evals README. by @gundermanc in
   [#20629](https://github.com/google-gemini/gemini-cli/pull/20629)
-- fix(core): skip telemetry logging for AbortError exceptions by @yunaseoul in
+* fix(core): skip telemetry logging for AbortError exceptions by @yunaseoul in
   [#19477](https://github.com/google-gemini/gemini-cli/pull/19477)
-- fix(core): restrict "System: Please continue" invalid stream retry to Gemini 2
+* fix(core): restrict "System: Please continue" invalid stream retry to Gemini 2
   models by @SandyTao520 in
   [#20897](https://github.com/google-gemini/gemini-cli/pull/20897)
-- ci(evals): only run evals in CI if prompts or tools changed by @gundermanc in
+* ci(evals): only run evals in CI if prompts or tools changed by @gundermanc in
   [#20898](https://github.com/google-gemini/gemini-cli/pull/20898)
-- Build binary by @aswinashok44 in
+* Build binary by @aswinashok44 in
   [#18933](https://github.com/google-gemini/gemini-cli/pull/18933)
-- Code review fixes as a pr by @jacob314 in
+* Code review fixes as a pr by @jacob314 in
   [#20612](https://github.com/google-gemini/gemini-cli/pull/20612)
-- fix(ci): handle empty APP_ID in stale PR closer by @bdmorgan in
+* fix(ci): handle empty APP_ID in stale PR closer by @bdmorgan in
   [#20919](https://github.com/google-gemini/gemini-cli/pull/20919)
-- feat(cli): invert context window display to show usage by @keithguerin in
+* feat(cli): invert context window display to show usage by @keithguerin in
   [#20071](https://github.com/google-gemini/gemini-cli/pull/20071)
-- fix(plan): clean up session directories and plans on deletion by @jerop in
+* fix(plan): clean up session directories and plans on deletion by @jerop in
   [#20914](https://github.com/google-gemini/gemini-cli/pull/20914)
-- fix(core): enforce optionality for API response fields in code_assist by
+* fix(core): enforce optionality for API response fields in code_assist by
   @sehoon38 in [#20714](https://github.com/google-gemini/gemini-cli/pull/20714)
-- feat(extensions): add support for plan directory in extension manifest by
+* feat(extensions): add support for plan directory in extension manifest by
   @mahimashanware in
   [#20354](https://github.com/google-gemini/gemini-cli/pull/20354)
-- feat(plan): enable built-in research subagents in plan mode by @Adib234 in
+* feat(plan): enable built-in research subagents in plan mode by @Adib234 in
   [#20972](https://github.com/google-gemini/gemini-cli/pull/20972)
-- feat(agents): directly indicate auth required state by @adamfweidman in
+* feat(agents): directly indicate auth required state by @adamfweidman in
   [#20986](https://github.com/google-gemini/gemini-cli/pull/20986)
-- fix(cli): wait for background auto-update before relaunching by @scidomino in
+* fix(cli): wait for background auto-update before relaunching by @scidomino in
   [#20904](https://github.com/google-gemini/gemini-cli/pull/20904)
-- fix: pre-load @scripts/copy_files.js references from external editor prompts
+* fix: pre-load @scripts/copy_files.js references from external editor prompts
   by @kartikangiras in
   [#20963](https://github.com/google-gemini/gemini-cli/pull/20963)
-- feat(evals): add behavioral evals for ask_user tool by @Adib234 in
+* feat(evals): add behavioral evals for ask_user tool by @Adib234 in
   [#20620](https://github.com/google-gemini/gemini-cli/pull/20620)
-- refactor common settings logic for skills,agents by @ishaanxgupta in
+* refactor common settings logic for skills,agents by @ishaanxgupta in
   [#17490](https://github.com/google-gemini/gemini-cli/pull/17490)
-- Update docs-writer skill with new resource by @g-samroberts in
+* Update docs-writer skill with new resource by @g-samroberts in
   [#20917](https://github.com/google-gemini/gemini-cli/pull/20917)
-- fix(cli): pin clipboardy to ~5.2.x by @scidomino in
+* fix(cli): pin clipboardy to ~5.2.x by @scidomino in
   [#21009](https://github.com/google-gemini/gemini-cli/pull/21009)
-- feat: Implement slash command handling in ACP for
+* feat: Implement slash command handling in ACP for
   `/memory`,`/init`,`/extensions` and `/restore` by @sripasg in
   [#20528](https://github.com/google-gemini/gemini-cli/pull/20528)
-- Docs/add hooks reference by @AadithyaAle in
+* Docs/add hooks reference by @AadithyaAle in
   [#20961](https://github.com/google-gemini/gemini-cli/pull/20961)
-- feat(plan): add copy subcommand to plan (#20491) by @ruomengz in
+* feat(plan): add copy subcommand to plan (#20491) by @ruomengz in
   [#20988](https://github.com/google-gemini/gemini-cli/pull/20988)
-- fix(core): sanitize and length-check MCP tool qualified names by @abhipatel12
+* fix(core): sanitize and length-check MCP tool qualified names by @abhipatel12
   in [#20987](https://github.com/google-gemini/gemini-cli/pull/20987)
-- Format the quota/limit style guide. by @g-samroberts in
+* Format the quota/limit style guide. by @g-samroberts in
   [#21017](https://github.com/google-gemini/gemini-cli/pull/21017)
-- fix(core): send shell output to model on cancel by @devr0306 in
+* fix(core): send shell output to model on cancel by @devr0306 in
   [#20501](https://github.com/google-gemini/gemini-cli/pull/20501)
-- remove hardcoded tiername when missing tier by @sehoon38 in
+* remove hardcoded tiername when missing tier by @sehoon38 in
   [#21022](https://github.com/google-gemini/gemini-cli/pull/21022)
-- feat(acp): add set models interface by @skeshive in
+* feat(acp): add set models interface by @skeshive in
   [#20991](https://github.com/google-gemini/gemini-cli/pull/20991)
 
 **Full Changelog**:
-https://github.com/google-gemini/gemini-cli/compare/v0.32.0-preview.0...v0.33.0-preview.4
+https://github.com/google-gemini/gemini-cli/compare/v0.32.0-preview.0...v0.33.0-preview.1

docs/cli/cli-reference.md

@@ -24,21 +24,6 @@ and parameters.
 | -------- | ----------------- | ------------------------------------------------------------------------------------------------------------------ |
 | `query`  | string (variadic) | Positional prompt. Defaults to one-shot mode. Use `-i/--prompt-interactive` to execute and continue interactively. |
 
-## Interactive commands
-
-These commands are available within the interactive REPL.
-
-| Command              | Description                              |
-| -------------------- | ---------------------------------------- |
-| `/skills reload`     | Reload discovered skills from disk       |
-| `/agents reload`     | Reload the agent registry                |
-| `/commands reload`   | Reload custom slash commands             |
-| `/memory reload`     | Reload context files (e.g., `GEMINI.md`) |
-| `/mcp reload`        | Restart and reload MCP servers           |
-| `/extensions reload` | Reload all active extensions             |
-| `/help`              | Show help for all commands               |
-| `/quit`              | Exit the interactive session             |
-
 ## CLI Options
 
 | Option                           | Alias | Type    | Default   | Description                                                                                                                                                            |

docs/cli/enterprise.md

@@ -244,7 +244,7 @@ gemini
 You can significantly enhance security by controlling which tools the Gemini
 model can use. This is achieved through the `tools.core` setting and the
 [Policy Engine](../reference/policy-engine.md). For a list of available tools,
-see the [Tools reference](../reference/tools.md).
+see the [Tools documentation](../tools/index.md).
 
 ### Allowlisting with `coreTools`
 
@@ -308,8 +308,8 @@ unintended tool execution.
 ## Managing custom tools (MCP servers)
 
 If your organization uses custom tools via
-[Model-Context Protocol (MCP) servers](../tools/mcp-server.md), it is crucial to
-understand how server configurations are managed to apply security policies
+[Model-Context Protocol (MCP) servers](../reference/tools-api.md), it is crucial
+to understand how server configurations are managed to apply security policies
 effectively.
 
 ### How MCP server configurations are merged

docs/cli/gemini-md.md

@@ -63,7 +63,7 @@ You can interact with the loaded context files by using the `/memory` command.
 - **`/memory show`**: Displays the full, concatenated content of the current
   hierarchical memory. This lets you inspect the exact instructional context
   being provided to the model.
-- **`/memory reload`**: Forces a re-scan and reload of all `GEMINI.md` files
+- **`/memory refresh`**: Forces a re-scan and reload of all `GEMINI.md` files
   from all configured locations.
 - **`/memory add <text>`**: Appends your text to your global
   `~/.gemini/GEMINI.md` file. This lets you add persistent memories on the fly.

docs/cli/notifications.md

@@ -1,58 +0,0 @@
-# Notifications (experimental)
-
-Gemini CLI can send system notifications to alert you when a session completes
-or when it needs your attention, such as when it's waiting for you to approve a
-tool call.
-
-> **Note:** This is a preview feature currently under active development.
-> Preview features may be available on the **Preview** channel or may need to be
-> enabled under `/settings`.
-
-Notifications are particularly useful when running long-running tasks or using
-[Plan Mode](./plan-mode.md), letting you switch to other windows while Gemini
-CLI works in the background.
-
-## Requirements
-
-Currently, system notifications are only supported on macOS.
-
-### Terminal support
-
-The CLI uses the OSC 9 terminal escape sequence to trigger system notifications.
-This is supported by several modern terminal emulators. If your terminal does
-not support OSC 9 notifications, Gemini CLI falls back to a system alert sound
-to get your attention.
-
-## Enable notifications
-
-Notifications are disabled by default. You can enable them using the `/settings`
-command or by updating your `settings.json` file.
-
-1.  Open the settings dialog by typing `/settings` in an interactive session.
-2.  Navigate to the **General** category.
-3.  Toggle the **Enable Notifications** setting to **On**.
-
-Alternatively, add the following to your `settings.json`:
-
-```json
-{
-  "general": {
-    "enableNotifications": true
-  }
-}
-```
-
-## Types of notifications
-
-Gemini CLI sends notifications for the following events:
-
-- **Action required:** Triggered when the model is waiting for user input or
-  tool approval. This helps you know when the CLI has paused and needs you to
-  intervene.
-- **Session complete:** Triggered when a session finishes successfully. This is
-  useful for tracking the completion of automated tasks.
-
-## Next steps
-
-- Start planning with [Plan Mode](./plan-mode.md).
-- Configure your experience with other [settings](./settings.md).

docs/cli/plan-mode.md

@@ -12,7 +12,8 @@ implementation. With Plan Mode, you can:
 > feedback is invaluable as we refine this feature. If you have ideas,
 > suggestions, or encounter issues:
 >
-> - [Open an issue] on GitHub.
+> - [Open an issue](https://github.com/google-gemini/gemini-cli/issues) on
+>   GitHub.
 > - Use the **/bug** command within Gemini CLI to file an issue.
 
 ## How to enable Plan Mode
@@ -131,10 +132,10 @@ These are the only allowed tools:
 
 ### Custom planning with skills
 
-You can use [Agent Skills] to customize how Gemini CLI approaches planning for
-specific types of tasks. When a skill is activated during Plan Mode, its
-specialized instructions and procedural workflows will guide the research,
-design, and planning phases.
+You can use [Agent Skills](./skills.md) to customize how Gemini CLI approaches
+planning for specific types of tasks. When a skill is activated during Plan
+Mode, its specialized instructions and procedural workflows will guide the
+research, design, and planning phases.
 
 For example:
 
@@ -251,59 +252,10 @@ modes = ["plan"]
 argsPattern = "\"file_path\":\"[^\"]+[\\\\/]+\\.gemini[\\\\/]+plans[\\\\/]+[\\w-]+\\.md\""
 ```
 
-## Planning workflows
-
-Plan Mode provides building blocks for structured research and design. These are
-implemented as [extensions] using core planning tools like [`enter_plan_mode`],
-[`exit_plan_mode`], and [`ask_user`].
-
-### Built-in planning workflow
-
-The built-in planner uses an adaptive workflow to analyze your project, consult
-you on trade-offs via [`ask_user`], and draft a plan for your approval.
-
-### Custom planning workflows
-
-You can install or create specialized planners to suit your workflow.
-
-#### Conductor
-
-[Conductor] is designed for spec-driven development. It organizes work into
-"tracks" and stores persistent artifacts in your project's `conductor/`
-directory:
-
-- **Automate transitions:** Switches to read-only mode via [`enter_plan_mode`].
-- **Streamline decisions:** Uses [`ask_user`] for architectural choices.
-- **Maintain project context:** Stores artifacts in the project directory using
-  [custom plan directory and policies](#custom-plan-directory-and-policies).
-- **Handoff execution:** Transitions to implementation via [`exit_plan_mode`].
-
-#### Build your own
-
-Since Plan Mode is built on modular building blocks, you can develop your own
-custom planning workflow as an [extensions]. By leveraging core tools and
-[custom policies](#custom-policies), you can define how Gemini CLI researches
-and stores plans for your specific domain.
-
-To build a custom planning workflow, you can use:
-
-- **Tool usage:** Use core tools like [`enter_plan_mode`], [`ask_user`], and
-  [`exit_plan_mode`] to manage the research and design process.
-- **Customization:** Set your own storage locations and policy rules using
-  [custom plan directories](#custom-plan-directory-and-policies) and
-  [custom policies](#custom-policies).
-
-> **Note:** Use [Conductor] as a reference when building your own custom
-> planning workflow.
-
-By using Plan Mode as its execution environment, your custom methodology can
-enforce read-only safety during the design phase while benefiting from
-high-reasoning model routing.
-
 ## Automatic Model Routing
 
-When using an [auto model], Gemini CLI automatically optimizes [model routing]
-based on the current phase of your task:
+When using an [**auto model**], Gemini CLI automatically optimizes [**model
+routing**] based on the current phase of your task:
 
 1.  **Planning Phase:** While in Plan Mode, the CLI routes requests to a
     high-reasoning **Pro** model to ensure robust architectural decisions and
@@ -354,8 +306,8 @@ those files are not automatically deleted and must be managed manually.
 [MCP tools]: /docs/tools/mcp-server.md
 [`save_memory`]: /docs/tools/memory.md
 [`activate_skill`]: /docs/cli/skills.md
-[`codebase_investigator`]: /docs/core/subagents.md#codebase-investigator
-[`cli_help`]: /docs/core/subagents.md#cli-help-agent
+[`codebase_investigator`]: /docs/core/subagents.md#codebase_investigator
+[`cli_help`]: /docs/core/subagents.md#cli_help
 [subagents]: /docs/core/subagents.md
 [custom subagents]: /docs/core/subagents.md#creating-custom-subagents
 [policy engine]: /docs/reference/policy-engine.md
@@ -365,11 +317,7 @@ those files are not automatically deleted and must be managed manually.
 [YOLO mode]: /docs/reference/configuration.md#command-line-arguments
 [`plan.toml`]:
   https://github.com/google-gemini/gemini-cli/blob/main/packages/core/src/policy/policies/plan.toml
-[auto model]: /docs/reference/configuration.md#model
+[auto model]: /docs/reference/configuration.md#model-settings
 [model routing]: /docs/cli/telemetry.md#model-routing
 [preferred external editor]: /docs/reference/configuration.md#general
 [session retention]: /docs/cli/session-management.md#session-retention
-[extensions]: /docs/extensions/
-[Conductor]: https://github.com/gemini-cli-extensions/conductor
-[open an issue]: https://github.com/google-gemini/gemini-cli/issues
-[Agent Skills]: /docs/cli/skills.md

docs/cli/sandbox.md

@@ -50,74 +50,6 @@ Cross-platform sandboxing with complete process isolation.
 **Note**: Requires building the sandbox image locally or using a published image
 from your organization's registry.
 
-### 3. gVisor / runsc (Linux only)
-
-Strongest isolation available: runs containers inside a user-space kernel via
-[gVisor](https://github.com/google/gvisor). gVisor intercepts all container
-system calls and handles them in a sandboxed kernel written in Go, providing a
-strong security barrier between AI operations and the host OS.
-
-**Prerequisites:**
-
-- Linux (gVisor supports Linux only)
-- Docker installed and running
-- gVisor/runsc runtime configured
-
-When you set `sandbox: "runsc"`, Gemini CLI runs
-`docker run --runtime=runsc ...` to execute containers with gVisor isolation.
-runsc is not auto-detected; you must specify it explicitly (e.g.
-`GEMINI_SANDBOX=runsc` or `sandbox: "runsc"`).
-
-To set up runsc:
-
-1.  Install the runsc binary.
-2.  Configure the Docker daemon to use the runsc runtime.
-3.  Verify the installation.
-
-### 4. LXC/LXD (Linux only, experimental)
-
-Full-system container sandboxing using LXC/LXD. Unlike Docker/Podman, LXC
-containers run a complete Linux system with `systemd`, `snapd`, and other system
-services. This is ideal for tools that don't work in standard Docker containers,
-such as Snapcraft and Rockcraft.
-
-**Prerequisites**:
-
-- Linux only.
-- LXC/LXD must be installed (`snap install lxd` or `apt install lxd`).
-- A container must be created and running before starting Gemini CLI. Gemini
-  does **not** create the container automatically.
-
-**Quick setup**:
-
-```bash
-# Initialize LXD (first time only)
-lxd init --auto
-
-# Create and start an Ubuntu container
-lxc launch ubuntu:24.04 gemini-sandbox
-
-# Enable LXC sandboxing
-export GEMINI_SANDBOX=lxc
-gemini -p "build the project"
-```
-
-**Custom container name**:
-
-```bash
-export GEMINI_SANDBOX=lxc
-export GEMINI_SANDBOX_IMAGE=my-snapcraft-container
-gemini -p "build the snap"
-```
-
-**Limitations**:
-
-- Linux only (LXC is not available on macOS or Windows).
-- The container must already exist and be running.
-- The workspace directory is bind-mounted into the container at the same
-  absolute path — the path must be writable inside the container.
-- Used with tools like Snapcraft or Rockcraft that require a full system.
-
 ## Quickstart
 
 ```bash
@@ -156,8 +88,7 @@ gemini -p "run the test suite"
 ### Enable sandboxing (in order of precedence)
 
 1. **Command flag**: `-s` or `--sandbox`
-2. **Environment variable**:
-   `GEMINI_SANDBOX=true|docker|podman|sandbox-exec|runsc|lxc`
+2. **Environment variable**: `GEMINI_SANDBOX=true|docker|podman|sandbox-exec`
 3. **Settings file**: `"sandbox": true` in the `tools` object of your
    `settings.json` file (e.g., `{"tools": {"sandbox": true}}`).
 

docs/cli/session-management.md

@@ -61,15 +61,6 @@ Browser**:
 /resume
 ```
 
-When typing `/resume` (or `/chat`) in slash completion, commands are grouped
-under titled separators:
-
-- `-- auto --` (session browser)
-  - `list` is selectable and opens the session browser
-- `-- checkpoints --` (manual tagged checkpoint commands)
-
-Unique prefixes such as `/resum` and `/cha` resolve to the same grouped menu.
-
 The Session Browser provides an interactive interface where you can perform the
 following actions:
 
@@ -81,21 +72,6 @@ following actions:
 - **Select:** Press **Enter** to resume the selected session.
 - **Esc:** Press **Esc** to exit the Session Browser.
 
-### Manual chat checkpoints
-
-For named branch points inside a session, use chat checkpoints:
-
-```text
-/resume save decision-point
-/resume list
-/resume resume decision-point
-```
-
-Compatibility aliases:
-
-- `/chat ...` works for the same commands.
-- `/resume checkpoints ...` also remains supported during migration.
-
 ## Managing sessions
 
 You can list and delete sessions to keep your history organized and manage disk

docs/cli/settings.md

@@ -57,7 +57,7 @@ they appear in the UI.
 | Show Shortcuts Hint                  | `ui.showShortcutsHint`                 | Show the "? for shortcuts" hint above the input.                                                                                                                  | `true`   |
 | Hide Banner                          | `ui.hideBanner`                        | Hide the application banner                                                                                                                                       | `false`  |
 | Hide Context Summary                 | `ui.hideContextSummary`                | Hide the context summary (GEMINI.md, MCP servers) above the input.                                                                                                | `false`  |
-| Hide CWD                             | `ui.footer.hideCWD`                    | Hide the current working directory in the footer.                                                                                                                 | `false`  |
+| Hide CWD                             | `ui.footer.hideCWD`                    | Hide the current working directory path in the footer.                                                                                                            | `false`  |
 | Hide Sandbox Status                  | `ui.footer.hideSandboxStatus`          | Hide the sandbox status indicator in the footer.                                                                                                                  | `false`  |
 | Hide Model Info                      | `ui.footer.hideModelInfo`              | Hide the model name and context usage in the footer.                                                                                                              | `false`  |
 | Hide Context Window Percentage       | `ui.footer.hideContextPercentage`      | Hides the context window usage percentage.                                                                                                                        | `true`   |
@@ -102,7 +102,7 @@ they appear in the UI.
 | UI Label                             | Setting                                           | Description                                                                                                                                                                                                                                 | Default |
 | ------------------------------------ | ------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------- |
 | Memory Discovery Max Dirs            | `context.discoveryMaxDirs`                        | Maximum number of directories to search for memory.                                                                                                                                                                                         | `200`   |
-| Load Memory From Include Directories | `context.loadMemoryFromIncludeDirectories`        | Controls how /memory reload loads GEMINI.md files. When true, include directories are scanned; when false, only the current directory is used.                                                                                              | `false` |
+| Load Memory From Include Directories | `context.loadMemoryFromIncludeDirectories`        | Controls how /memory refresh loads GEMINI.md files. When true, include directories are scanned; when false, only the current directory is used.                                                                                             | `false` |
 | Respect .gitignore                   | `context.fileFiltering.respectGitIgnore`          | Respect .gitignore files when searching.                                                                                                                                                                                                    | `true`  |
 | Respect .geminiignore                | `context.fileFiltering.respectGeminiIgnore`       | Respect .geminiignore files when searching.                                                                                                                                                                                                 | `true`  |
 | Enable Recursive File Search         | `context.fileFiltering.enableRecursiveFileSearch` | Enable recursive file search functionality when completing @ references in the prompt.                                                                                                                                                      | `true`  |
@@ -147,6 +147,7 @@ they appear in the UI.
 | Plan                       | `experimental.plan`                      | Enable planning features (Plan Mode and tools).                                                                                                           | `false` |
 | Model Steering             | `experimental.modelSteering`             | Enable model steering (user hints) to guide the model during tool execution.                                                                              | `false` |
 | Direct Web Fetch           | `experimental.directWebFetch`            | Enable web fetch behavior that bypasses LLM summarization.                                                                                                | `false` |
+| Enable Gemma Model Router  | `experimental.gemmaModelRouter.enabled`  | Enable the Gemma Model Router. Requires a local endpoint serving Gemma via the Gemini API using LiteRT-LM shim.                                           | `false` |
 
 ### Skills
 

docs/cli/tutorials/mcp-setup.md

@@ -101,8 +101,8 @@ The agent will:
 
 - **Server won't start?** Try running the docker command manually in your
   terminal to see if it prints an error (e.g., "image not found").
-- **Tools not found?** Run `/mcp reload` to force the CLI to re-query the server
-  for its capabilities.
+- **Tools not found?** Run `/mcp refresh` to force the CLI to re-query the
+  server for its capabilities.
 
 ## Next steps
 

docs/cli/tutorials/memory-management.md

@@ -105,7 +105,7 @@ excellent for debugging why the agent might be ignoring a rule.
 If you edit a `GEMINI.md` file while a session is running, the agent won't know
 immediately. Force a reload with:
 
-**Command:** `/memory reload`
+**Command:** `/memory refresh`
 
 ## Best practices
 

docs/cli/tutorials/session-management.md

@@ -89,9 +89,9 @@ Gemini gives you granular control over the undo process. You can choose to:
 Sometimes you want to try two different approaches to the same problem.
 
 1.  Start a session and get to a decision point.
-2.  Save the current state with `/resume save decision-point`.
+2.  Save the current state with `/chat save decision-point`.
 3.  Try your first approach.
-4.  Later, use `/resume resume decision-point` to fork the conversation back to
+4.  Later, use `/chat resume decision-point` to fork the conversation back to
     that moment and try a different approach.
 
 This creates a new branch of history without losing your original work.
@@ -101,5 +101,5 @@ This creates a new branch of history without losing your original work.
 - Learn about [Checkpointing](../../cli/checkpointing.md) to understand the
   underlying safety mechanism.
 - Explore [Task planning](task-planning.md) to keep complex sessions organized.
-- See the [Command reference](../../reference/commands.md) for `/resume`
-  options, grouped checkpoint menus, and `/chat` compatibility aliases.
+- See the [Command reference](../../reference/commands.md) for all `/chat` and
+  `/resume` options.

docs/cli/tutorials/shell-commands.md

@@ -17,10 +17,9 @@ prefix.
 
 **Example:** `!ls -la`
 
-This executes `ls -la` immediately and prints the output to your terminal.
-Gemini CLI also records the command and its output in the current session
-context, so the model can reference it in follow-up prompts. Very large outputs
-may be truncated.
+This executes `ls -la` immediately and prints the output to your terminal. The
+AI doesn't "see" this output unless you paste it back into the chat or use it in
+a prompt.
 
 ### Scenario: Entering Shell mode
 

docs/core/index.md

@@ -9,8 +9,8 @@ requests sent from `packages/cli`. For a general overview of Gemini CLI, see the
 
 - **[Sub-agents (experimental)](./subagents.md):** Learn how to create and use
   specialized sub-agents for complex tasks.
-- **[Core tools reference](../reference/tools.md):** Information on how tools
-  are defined, registered, and used by the core.
+- **[Core tools API](../reference/tools-api.md):** Information on how tools are
+  defined, registered, and used by the core.
 - **[Memory Import Processor](../reference/memport.md):** Documentation for the
   modular GEMINI.md import feature using @file.md syntax.
 - **[Policy Engine](../reference/policy-engine.md):** Use the Policy Engine for

docs/core/remote-agents.md

@@ -75,7 +75,7 @@ Markdown file.
 Users can manage subagents using the following commands within the Gemini CLI:
 
 - `/agents list`: Displays all available local and remote subagents.
-- `/agents reload`: Reloads the agent registry. Use this after adding or
+- `/agents refresh`: Reloads the agent registry. Use this after adding or
   modifying agent definition files.
 - `/agents enable <agent_name>`: Enables a specific subagent.
 - `/agents disable <agent_name>`: Disables a specific subagent.

docs/get-started/authentication.md

@@ -4,10 +4,6 @@ To use Gemini CLI, you'll need to authenticate with Google. This guide helps you
 quickly find the best way to sign in based on your account type and how you're
 using the CLI.
 
-> **Note:** Looking for a high-level comparison of all available subscriptions?
-> To compare features and find the right quota for your needs, see our
-> [Plans page](/plans/).
-
 For most users, we recommend starting Gemini CLI and logging in with your
 personal Google account.
 

docs/get-started/gemini-3.md

@@ -39,10 +39,6 @@ When you encounter that limit, you’ll be given the option to switch to Gemini
 2.5 Pro, upgrade for higher limits, or stop. You’ll also be told when your usage
 limit resets and Gemini 3 Pro can be used again.
 
-> **Note:** Looking to upgrade for higher limits? To compare subscription
-> options and find the right quota for your needs, see our
-> [Plans page](/plans/).
-
 Similarly, when you reach your daily usage limit for Gemini 2.5 Pro, you’ll see
 a message prompting fallback to Gemini 2.5 Flash.
 

docs/get-started/index.md

@@ -72,7 +72,7 @@ session's token usage, as well as your overall quota and usage for the supported
 models.
 
 For more information on the `/stats` command and its subcommands, see the
-[Command Reference](../reference/commands.md#stats).
+[Command Reference](../../reference/commands.md#stats).
 
 ## Next steps
 

docs/hooks/reference.md

@@ -82,8 +82,8 @@ For `BeforeTool` and `AfterTool` events, the `matcher` field in your settings is
 compared against the name of the tool being executed.
 
 - **Built-in Tools**: You can match any built-in tool (e.g., `read_file`,
-  `run_shell_command`). See the [Tools Reference](/docs/reference/tools) for a
-  full list of available tool names.
+  `run_shell_command`). See the [Tools Reference](/docs/tools) for a full list
+  of available tool names.
 - **MCP Tools**: Tools from MCP servers follow the naming pattern
   `mcp__<server_name>__<tool_name>`.
 - **Regex Support**: Matchers support regular expressions (e.g.,

docs/index.md

@@ -108,8 +108,8 @@ Deep technical documentation and API specifications.
   processes memory from various sources.
 - **[Policy engine](./reference/policy-engine.md):** Fine-grained execution
   control.
-- **[Tools reference](./reference/tools.md):** Information on how tools are
-  defined, registered, and used.
+- **[Tools API](./reference/tools-api.md):** The API for defining and using
+  tools.
 
 ## Resources
 

docs/issue-and-pr-automation.md

@@ -113,45 +113,7 @@ process.
     ensure every issue is eventually categorized, even if the initial triage
     fails.
 
-### 5. Automatic unassignment of inactive contributors: `Unassign Inactive Issue Assignees`
-
-To keep the list of open `help wanted` issues accessible to all contributors,
-this workflow automatically removes **external contributors** who have not
-opened a linked pull request within **7 days** of being assigned. Maintainers,
-org members, and repo collaborators with write access or above are always exempt
-and will never be auto-unassigned.
-
-- **Workflow File**: `.github/workflows/unassign-inactive-assignees.yml`
-- **When it runs**: Every day at 09:00 UTC, and can be triggered manually with
-  an optional `dry_run` mode.
-- **What it does**:
-  1. Finds every open issue labeled `help wanted` that has at least one
-     assignee.
-  2. Identifies privileged users (team members, repo collaborators with write+
-     access, maintainers) and skips them entirely.
-  3. For each remaining (external) assignee it reads the issue's timeline to
-     determine:
-     - The exact date they were assigned (using `assigned` timeline events).
-     - Whether they have opened a PR that is already linked/cross-referenced to
-       the issue.
-  4. Each cross-referenced PR is fetched to verify it is **ready for review**:
-     open and non-draft, or already merged. Draft PRs do not count.
-  5. If an assignee has been assigned for **more than 7 days** and no qualifying
-     PR is found, they are automatically unassigned and a comment is posted
-     explaining the reason and how to re-claim the issue.
-  6. Assignees who have a non-draft, open or merged PR linked to the issue are
-     **never** unassigned by this workflow.
-- **What you should do**:
-  - **Open a real PR, not a draft**: Within 7 days of being assigned, open a PR
-    that is ready for review and include `Fixes #<issue-number>` in the
-    description. Draft PRs do not satisfy the requirement and will not prevent
-    auto-unassignment.
-  - **Re-assign if unassigned by mistake**: Comment `/assign` on the issue to
-    assign yourself again.
-  - **Unassign yourself** if you can no longer work on the issue by commenting
-    `/unassign`, so other contributors can pick it up right away.
-
-### 6. Release automation
+### 5. Release automation
 
 This workflow handles the process of packaging and publishing new versions of
 the Gemini CLI.

docs/redirects.json

@@ -8,8 +8,7 @@
   "/docs/core/concepts": "/docs",
   "/docs/core/memport": "/docs/reference/memport",
   "/docs/core/policy-engine": "/docs/reference/policy-engine",
-  "/docs/core/tools-api": "/docs/reference/tools",
-  "/docs/reference/tools-api": "/docs/reference/tools",
+  "/docs/core/tools-api": "/docs/reference/tools-api",
   "/docs/faq": "/docs/resources/faq",
   "/docs/get-started/configuration": "/docs/reference/configuration",
   "/docs/get-started/configuration-v1": "/docs/reference/configuration",

docs/reference/commands.md

@@ -28,33 +28,24 @@ Slash commands provide meta-level control over the CLI itself.
 
 ### `/chat`
 
-- **Description:** Alias for `/resume`. Both commands now expose the same
-  session browser action and checkpoint subcommands.
-- **Menu layout when typing `/chat` (or `/resume`)**:
-  - `-- auto --`
-    - `list` (selecting this opens the auto-saved session browser)
-  - `-- checkpoints --`
-    - `list`, `save`, `resume`, `delete`, `share` (manual tagged checkpoints)
-  - **Note:** Unique prefixes (for example `/cha` or `/resum`) resolve to the
-    same grouped menu.
+- **Description:** Save and resume conversation history for branching
+  conversation state interactively, or resuming a previous state from a later
+  session.
 - **Sub-commands:**
   - **`debug`**
     - **Description:** Export the most recent API request as a JSON payload.
   - **`delete <tag>`**
     - **Description:** Deletes a saved conversation checkpoint.
-    - **Equivalent:** `/resume delete <tag>`
   - **`list`**
-    - **Description:** Lists available tags for manually saved checkpoints.
+    - **Description:** Lists available tags for chat state resumption.
     - **Note:** This command only lists chats saved within the current project.
       Because chat history is project-scoped, chats saved in other project
       directories will not be displayed.
-    - **Equivalent:** `/resume list`
   - **`resume <tag>`**
     - **Description:** Resumes a conversation from a previous save.
     - **Note:** You can only resume chats that were saved within the current
       project. To resume a chat from a different project, you must run the
       Gemini CLI from that project's directory.
-    - **Equivalent:** `/resume resume <tag>`
   - **`save <tag>`**
     - **Description:** Saves the current conversation history. You must add a
       `<tag>` for identifying the conversation state.
@@ -69,12 +60,10 @@ Slash commands provide meta-level control over the CLI itself.
         conversation states. For automatic checkpoints created before file
         modifications, see the
         [Checkpointing documentation](../cli/checkpointing.md).
-      - **Equivalent:** `/resume save <tag>`
   - **`share [filename]`**
     - **Description** Writes the current conversation to a provided Markdown or
       JSON file. If no filename is provided, then the CLI will generate one.
     - **Usage** `/chat share file.md` or `/chat share file.json`.
-    - **Equivalent:** `/resume share [filename]`
 
 ### `/clear`
 
@@ -325,13 +314,10 @@ Slash commands provide meta-level control over the CLI itself.
 
 ### `/resume`
 
-- **Description:** Browse and resume previous conversation sessions, and manage
-  manual chat checkpoints.
+- **Description:** Browse and resume previous conversation sessions. Opens an
+  interactive session browser where you can search, filter, and select from
+  automatically saved conversations.
 - **Features:**
-  - **Auto sessions:** Run `/resume` to open the interactive session browser for
-    automatically saved conversations.
-  - **Chat checkpoints:** Use checkpoint subcommands directly (`/resume save`,
-    `/resume resume`, etc.).
   - **Management:** Delete unwanted sessions directly from the browser
   - **Resume:** Select any session to resume and continue the conversation
   - **Search:** Use `/` to search through conversation content across all
@@ -342,23 +328,6 @@ Slash commands provide meta-level control over the CLI itself.
 - **Note:** All conversations are automatically saved as you chat - no manual
   saving required. See [Session Management](../cli/session-management.md) for
   complete details.
-- **Alias:** `/chat` provides the same behavior and subcommands.
-- **Sub-commands:**
-  - **`list`**
-    - **Description:** Lists available tags for manual chat checkpoints.
-  - **`save <tag>`**
-    - **Description:** Saves the current conversation as a tagged checkpoint.
-  - **`resume <tag>`** (alias: `load`)
-    - **Description:** Loads a previously saved tagged checkpoint.
-  - **`delete <tag>`**
-    - **Description:** Deletes a tagged checkpoint.
-  - **`share [filename]`**
-    - **Description:** Exports the current conversation to Markdown or JSON.
-  - **`debug`**
-    - **Description:** Export the most recent API request as JSON payload
-      (nightly builds).
-  - **Compatibility alias:** `/resume checkpoints ...` is still accepted for the
-    same checkpoint commands.
 
 ### `/settings`
 

docs/reference/configuration.md

@@ -250,18 +250,8 @@ their corresponding top-level category object in your `settings.json` file.
     input.
   - **Default:** `false`
 
-- **`ui.footer.items`** (array):
-  - **Description:** List of item IDs to display in the footer. Rendered in
-    order
-  - **Default:** `undefined`
-
-- **`ui.footer.showLabels`** (boolean):
-  - **Description:** Display a second line above the footer items with
-    descriptive headers (e.g., /model).
-  - **Default:** `true`
-
 - **`ui.footer.hideCWD`** (boolean):
-  - **Description:** Hide the current working directory in the footer.
+  - **Description:** Hide the current working directory path in the footer.
   - **Default:** `false`
 
 - **`ui.footer.hideSandboxStatus`** (boolean):
@@ -719,7 +709,7 @@ their corresponding top-level category object in your `settings.json` file.
   - **Default:** `[]`
 
 - **`context.loadMemoryFromIncludeDirectories`** (boolean):
-  - **Description:** Controls how /memory reload loads GEMINI.md files. When
+  - **Description:** Controls how /memory refresh loads GEMINI.md files. When
     true, include directories are scanned; when false, only the current
     directory is used.
   - **Default:** `false`
@@ -757,8 +747,7 @@ their corresponding top-level category object in your `settings.json` file.
 
 - **`tools.sandbox`** (boolean | string):
   - **Description:** Sandbox execution environment. Set to a boolean to enable
-    or disable the sandbox, provide a string path to a sandbox profile, or
-    specify an explicit sandbox command (e.g., "docker", "podman", "lxc").
+    or disable the sandbox, or provide a string path to a sandbox profile.
   - **Default:** `undefined`
   - **Requires restart:** Yes
 
@@ -1041,8 +1030,8 @@ their corresponding top-level category object in your `settings.json` file.
   - **Requires restart:** Yes
 
 - **`experimental.gemmaModelRouter.enabled`** (boolean):
-  - **Description:** Enable the Gemma Model Router (experimental). Requires a
-    local endpoint serving Gemma via the Gemini API using LiteRT-LM shim.
+  - **Description:** Enable the Gemma Model Router. Requires a local endpoint
+    serving Gemma via the Gemini API using LiteRT-LM shim.
   - **Default:** `false`
   - **Requires restart:** Yes
 
@@ -1705,7 +1694,7 @@ conventions and context.
     loaded, allowing you to verify the hierarchy and content being used by the
     AI.
   - See the [Commands documentation](./commands.md#memory) for full details on
-    the `/memory` command and its sub-commands (`show` and `reload`).
+    the `/memory` command and its sub-commands (`show` and `refresh`).
 
 By understanding and utilizing these configuration layers and the hierarchical
 nature of context files, you can effectively manage the AI's memory and tailor

docs/reference/keyboard-shortcuts.md

@@ -8,119 +8,119 @@ available combinations.
 
 #### Basic Controls
 
-| Action                                                          | Keys                |
-| --------------------------------------------------------------- | ------------------- |
-| Confirm the current selection or choice.                        | `Enter`             |
-| Dismiss dialogs or cancel the current focus.                    | `Esc`<br />`Ctrl+[` |
-| Cancel the current request or quit the CLI when input is empty. | `Ctrl+C`            |
-| Exit the CLI when the input buffer is empty.                    | `Ctrl+D`            |
+| Action                                                          | Keys                  |
+| --------------------------------------------------------------- | --------------------- |
+| Confirm the current selection or choice.                        | `Enter`               |
+| Dismiss dialogs or cancel the current focus.                    | `Esc`<br />`Ctrl + [` |
+| Cancel the current request or quit the CLI when input is empty. | `Ctrl + C`            |
+| Exit the CLI when the input buffer is empty.                    | `Ctrl + D`            |
 
 #### Cursor Movement
 
-| Action                                      | Keys                                       |
-| ------------------------------------------- | ------------------------------------------ |
-| Move the cursor to the start of the line.   | `Ctrl+A`<br />`Home`                       |
-| Move the cursor to the end of the line.     | `Ctrl+E`<br />`End`                        |
-| Move the cursor up one line.                | `Up`                                       |
-| Move the cursor down one line.              | `Down`                                     |
-| Move the cursor one character to the left.  | `Left`                                     |
-| Move the cursor one character to the right. | `Right`<br />`Ctrl+F`                      |
-| Move the cursor one word to the left.       | `Ctrl+Left`<br />`Alt+Left`<br />`Alt+B`   |
-| Move the cursor one word to the right.      | `Ctrl+Right`<br />`Alt+Right`<br />`Alt+F` |
+| Action                                      | Keys                                                         |
+| ------------------------------------------- | ------------------------------------------------------------ |
+| Move the cursor to the start of the line.   | `Ctrl + A`<br />`Home (no Shift, Ctrl)`                      |
+| Move the cursor to the end of the line.     | `Ctrl + E`<br />`End (no Shift, Ctrl)`                       |
+| Move the cursor up one line.                | `Up Arrow (no Shift, Alt, Ctrl, Cmd)`                        |
+| Move the cursor down one line.              | `Down Arrow (no Shift, Alt, Ctrl, Cmd)`                      |
+| Move the cursor one character to the left.  | `Left Arrow (no Shift, Alt, Ctrl, Cmd)`                      |
+| Move the cursor one character to the right. | `Right Arrow (no Shift, Alt, Ctrl, Cmd)`<br />`Ctrl + F`     |
+| Move the cursor one word to the left.       | `Ctrl + Left Arrow`<br />`Alt + Left Arrow`<br />`Alt + B`   |
+| Move the cursor one word to the right.      | `Ctrl + Right Arrow`<br />`Alt + Right Arrow`<br />`Alt + F` |
 
 #### Editing
 
-| Action                                           | Keys                                                     |
-| ------------------------------------------------ | -------------------------------------------------------- |
-| Delete from the cursor to the end of the line.   | `Ctrl+K`                                                 |
-| Delete from the cursor to the start of the line. | `Ctrl+U`                                                 |
-| Clear all text in the input field.               | `Ctrl+C`                                                 |
-| Delete the previous word.                        | `Ctrl+Backspace`<br />`Alt+Backspace`<br />`Ctrl+W`      |
-| Delete the next word.                            | `Ctrl+Delete`<br />`Alt+Delete`<br />`Alt+D`             |
-| Delete the character to the left.                | `Backspace`<br />`Ctrl+H`                                |
-| Delete the character to the right.               | `Delete`<br />`Ctrl+D`                                   |
-| Undo the most recent text edit.                  | `Cmd/Win+Z`<br />`Alt+Z`                                 |
-| Redo the most recent undone text edit.           | `Ctrl+Shift+Z`<br />`Shift+Cmd/Win+Z`<br />`Alt+Shift+Z` |
+| Action                                           | Keys                                                             |
+| ------------------------------------------------ | ---------------------------------------------------------------- |
+| Delete from the cursor to the end of the line.   | `Ctrl + K`                                                       |
+| Delete from the cursor to the start of the line. | `Ctrl + U`                                                       |
+| Clear all text in the input field.               | `Ctrl + C`                                                       |
+| Delete the previous word.                        | `Ctrl + Backspace`<br />`Alt + Backspace`<br />`Ctrl + W`        |
+| Delete the next word.                            | `Ctrl + Delete`<br />`Alt + Delete`<br />`Alt + D`               |
+| Delete the character to the left.                | `Backspace`<br />`Ctrl + H`                                      |
+| Delete the character to the right.               | `Delete`<br />`Ctrl + D`                                         |
+| Undo the most recent text edit.                  | `Cmd + Z (no Shift)`<br />`Alt + Z (no Shift)`                   |
+| Redo the most recent undone text edit.           | `Shift + Ctrl + Z`<br />`Shift + Cmd + Z`<br />`Shift + Alt + Z` |
 
 #### Scrolling
 
-| Action                   | Keys                          |
-| ------------------------ | ----------------------------- |
-| Scroll content up.       | `Shift+Up`                    |
-| Scroll content down.     | `Shift+Down`                  |
-| Scroll to the top.       | `Ctrl+Home`<br />`Shift+Home` |
-| Scroll to the bottom.    | `Ctrl+End`<br />`Shift+End`   |
-| Scroll up by one page.   | `Page Up`                     |
-| Scroll down by one page. | `Page Down`                   |
+| Action                   | Keys                              |
+| ------------------------ | --------------------------------- |
+| Scroll content up.       | `Shift + Up Arrow`                |
+| Scroll content down.     | `Shift + Down Arrow`              |
+| Scroll to the top.       | `Ctrl + Home`<br />`Shift + Home` |
+| Scroll to the bottom.    | `Ctrl + End`<br />`Shift + End`   |
+| Scroll up by one page.   | `Page Up`                         |
+| Scroll down by one page. | `Page Down`                       |
 
 #### History & Search
 
-| Action                                       | Keys         |
-| -------------------------------------------- | ------------ |
-| Show the previous entry in history.          | `Ctrl+P`     |
-| Show the next entry in history.              | `Ctrl+N`     |
-| Start reverse search through history.        | `Ctrl+R`     |
-| Submit the selected reverse-search match.    | `Enter`      |
-| Accept a suggestion while reverse searching. | `Tab`        |
-| Browse and rewind previous interactions.     | `Double Esc` |
+| Action                                       | Keys                  |
+| -------------------------------------------- | --------------------- |
+| Show the previous entry in history.          | `Ctrl + P (no Shift)` |
+| Show the next entry in history.              | `Ctrl + N (no Shift)` |
+| Start reverse search through history.        | `Ctrl + R`            |
+| Submit the selected reverse-search match.    | `Enter (no Ctrl)`     |
+| Accept a suggestion while reverse searching. | `Tab (no Shift)`      |
+| Browse and rewind previous interactions.     | `Double Esc`          |
 
 #### Navigation
 
-| Action                                             | Keys            |
-| -------------------------------------------------- | --------------- |
-| Move selection up in lists.                        | `Up`            |
-| Move selection down in lists.                      | `Down`          |
-| Move up within dialog options.                     | `Up`<br />`K`   |
-| Move down within dialog options.                   | `Down`<br />`J` |
-| Move to the next item or question in a dialog.     | `Tab`           |
-| Move to the previous item or question in a dialog. | `Shift+Tab`     |
+| Action                                             | Keys                                        |
+| -------------------------------------------------- | ------------------------------------------- |
+| Move selection up in lists.                        | `Up Arrow (no Shift)`                       |
+| Move selection down in lists.                      | `Down Arrow (no Shift)`                     |
+| Move up within dialog options.                     | `Up Arrow (no Shift)`<br />`K (no Shift)`   |
+| Move down within dialog options.                   | `Down Arrow (no Shift)`<br />`J (no Shift)` |
+| Move to the next item or question in a dialog.     | `Tab (no Shift)`                            |
+| Move to the previous item or question in a dialog. | `Shift + Tab`                               |
 
 #### Suggestions & Completions
 
-| Action                                  | Keys                 |
-| --------------------------------------- | -------------------- |
-| Accept the inline suggestion.           | `Tab`<br />`Enter`   |
-| Move to the previous completion option. | `Up`<br />`Ctrl+P`   |
-| Move to the next completion option.     | `Down`<br />`Ctrl+N` |
-| Expand an inline suggestion.            | `Right`              |
-| Collapse an inline suggestion.          | `Left`               |
+| Action                                  | Keys                                               |
+| --------------------------------------- | -------------------------------------------------- |
+| Accept the inline suggestion.           | `Tab (no Shift)`<br />`Enter (no Ctrl)`            |
+| Move to the previous completion option. | `Up Arrow (no Shift)`<br />`Ctrl + P (no Shift)`   |
+| Move to the next completion option.     | `Down Arrow (no Shift)`<br />`Ctrl + N (no Shift)` |
+| Expand an inline suggestion.            | `Right Arrow`                                      |
+| Collapse an inline suggestion.          | `Left Arrow`                                       |
 
 #### Text Input
 
-| Action                                                     | Keys                                                                                |
-| ---------------------------------------------------------- | ----------------------------------------------------------------------------------- |
-| Submit the current prompt.                                 | `Enter`                                                                             |
-| Insert a newline without submitting.                       | `Ctrl+Enter`<br />`Cmd/Win+Enter`<br />`Alt+Enter`<br />`Shift+Enter`<br />`Ctrl+J` |
-| Open the current prompt or the plan in an external editor. | `Ctrl+X`                                                                            |
-| Paste from the clipboard.                                  | `Ctrl+V`<br />`Cmd/Win+V`<br />`Alt+V`                                              |
+| Action                                                     | Keys                                                                                      |
+| ---------------------------------------------------------- | ----------------------------------------------------------------------------------------- |
+| Submit the current prompt.                                 | `Enter (no Shift, Alt, Ctrl, Cmd)`                                                        |
+| Insert a newline without submitting.                       | `Ctrl + Enter`<br />`Cmd + Enter`<br />`Alt + Enter`<br />`Shift + Enter`<br />`Ctrl + J` |
+| Open the current prompt or the plan in an external editor. | `Ctrl + X`                                                                                |
+| Paste from the clipboard.                                  | `Ctrl + V`<br />`Cmd + V`<br />`Alt + V`                                                  |
 
 #### App Controls
 
-| Action                                                                                                                                             | Keys               |
-| -------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------ |
-| Toggle detailed error information.                                                                                                                 | `F12`              |
-| Toggle the full TODO list.                                                                                                                         | `Ctrl+T`           |
-| Show IDE context details.                                                                                                                          | `Ctrl+G`           |
-| Toggle Markdown rendering.                                                                                                                         | `Alt+M`            |
-| Toggle copy mode when in alternate buffer mode.                                                                                                    | `Ctrl+S`           |
-| Toggle YOLO (auto-approval) mode for tool calls.                                                                                                   | `Ctrl+Y`           |
-| Cycle through approval modes: default (prompt), auto_edit (auto-approve edits), and plan (read-only). Plan mode is skipped when the agent is busy. | `Shift+Tab`        |
-| Expand and collapse blocks of content when not in alternate buffer mode.                                                                           | `Ctrl+O`           |
-| Expand or collapse a paste placeholder when cursor is over placeholder.                                                                            | `Ctrl+O`           |
-| Toggle current background shell visibility.                                                                                                        | `Ctrl+B`           |
-| Toggle background shell list.                                                                                                                      | `Ctrl+L`           |
-| Kill the active background shell.                                                                                                                  | `Ctrl+K`           |
-| Confirm selection in background shell list.                                                                                                        | `Enter`            |
-| Dismiss background shell list.                                                                                                                     | `Esc`              |
-| Move focus from background shell to Gemini.                                                                                                        | `Shift+Tab`        |
-| Move focus from background shell list to Gemini.                                                                                                   | `Tab`              |
-| Show warning when trying to move focus away from background shell.                                                                                 | `Tab`              |
-| Show warning when trying to move focus away from shell input.                                                                                      | `Tab`              |
-| Move focus from Gemini to the active shell.                                                                                                        | `Tab`              |
-| Move focus from the shell back to Gemini.                                                                                                          | `Shift+Tab`        |
-| Clear the terminal screen and redraw the UI.                                                                                                       | `Ctrl+L`           |
-| Restart the application.                                                                                                                           | `R`<br />`Shift+R` |
-| Suspend the CLI and move it to the background.                                                                                                     | `Ctrl+Z`           |
+| Action                                                                                                                                             | Keys             |
+| -------------------------------------------------------------------------------------------------------------------------------------------------- | ---------------- |
+| Toggle detailed error information.                                                                                                                 | `F12`            |
+| Toggle the full TODO list.                                                                                                                         | `Ctrl + T`       |
+| Show IDE context details.                                                                                                                          | `Ctrl + G`       |
+| Toggle Markdown rendering.                                                                                                                         | `Alt + M`        |
+| Toggle copy mode when in alternate buffer mode.                                                                                                    | `Ctrl + S`       |
+| Toggle YOLO (auto-approval) mode for tool calls.                                                                                                   | `Ctrl + Y`       |
+| Cycle through approval modes: default (prompt), auto_edit (auto-approve edits), and plan (read-only). Plan mode is skipped when the agent is busy. | `Shift + Tab`    |
+| Expand and collapse blocks of content when not in alternate buffer mode.                                                                           | `Ctrl + O`       |
+| Expand or collapse a paste placeholder when cursor is over placeholder.                                                                            | `Ctrl + O`       |
+| Toggle current background shell visibility.                                                                                                        | `Ctrl + B`       |
+| Toggle background shell list.                                                                                                                      | `Ctrl + L`       |
+| Kill the active background shell.                                                                                                                  | `Ctrl + K`       |
+| Confirm selection in background shell list.                                                                                                        | `Enter`          |
+| Dismiss background shell list.                                                                                                                     | `Esc`            |
+| Move focus from background shell to Gemini.                                                                                                        | `Shift + Tab`    |
+| Move focus from background shell list to Gemini.                                                                                                   | `Tab (no Shift)` |
+| Show warning when trying to move focus away from background shell.                                                                                 | `Tab (no Shift)` |
+| Show warning when trying to move focus away from shell input.                                                                                      | `Tab (no Shift)` |
+| Move focus from Gemini to the active shell.                                                                                                        | `Tab (no Shift)` |
+| Move focus from the shell back to Gemini.                                                                                                          | `Shift + Tab`    |
+| Clear the terminal screen and redraw the UI.                                                                                                       | `Ctrl + L`       |
+| Restart the application.                                                                                                                           | `R`              |
+| Suspend the CLI and move it to the background.                                                                                                     | `Ctrl + Z`       |
 
 <!-- KEYBINDINGS-AUTOGEN:END -->
 
@@ -156,7 +156,7 @@ available combinations.
 ## Limitations
 
 - On [Windows Terminal](https://en.wikipedia.org/wiki/Windows_Terminal):
-  - `shift+enter` is only supported in version 1.25 and higher.
+  - `shift+enter` is not supported.
   - `shift+tab`
     [is not supported](https://github.com/google-gemini/gemini-cli/issues/20314)
     on Node 20 and earlier versions of Node 22.

docs/reference/tools-api.md

@@ -0,0 +1,131 @@
+# Gemini CLI core: Tools API
+
+The Gemini CLI core (`packages/core`) features a robust system for defining,
+registering, and executing tools. These tools extend the capabilities of the
+Gemini model, allowing it to interact with the local environment, fetch web
+content, and perform various actions beyond simple text generation.
+
+## Core concepts
+
+- **Tool (`tools.ts`):** An interface and base class (`BaseTool`) that defines
+  the contract for all tools. Each tool must have:
+  - `name`: A unique internal name (used in API calls to Gemini).
+  - `displayName`: A user-friendly name.
+  - `description`: A clear explanation of what the tool does, which is provided
+    to the Gemini model.
+  - `parameterSchema`: A JSON schema defining the parameters that the tool
+    accepts. This is crucial for the Gemini model to understand how to call the
+    tool correctly.
+  - `validateToolParams()`: A method to validate incoming parameters.
+  - `getDescription()`: A method to provide a human-readable description of what
+    the tool will do with specific parameters before execution.
+  - `shouldConfirmExecute()`: A method to determine if user confirmation is
+    required before execution (e.g., for potentially destructive operations).
+  - `execute()`: The core method that performs the tool's action and returns a
+    `ToolResult`.
+
+- **`ToolResult` (`tools.ts`):** An interface defining the structure of a tool's
+  execution outcome:
+  - `llmContent`: The factual content to be included in the history sent back to
+    the LLM for context. This can be a simple string or a `PartListUnion` (an
+    array of `Part` objects and strings) for rich content.
+  - `returnDisplay`: A user-friendly string (often Markdown) or a special object
+    (like `FileDiff`) for display in the CLI.
+
+- **Returning rich content:** Tools are not limited to returning simple text.
+  The `llmContent` can be a `PartListUnion`, which is an array that can contain
+  a mix of `Part` objects (for images, audio, etc.) and `string`s. This allows a
+  single tool execution to return multiple pieces of rich content.
+
+- **Tool registry (`tool-registry.ts`):** A class (`ToolRegistry`) responsible
+  for:
+  - **Registering tools:** Holding a collection of all available built-in tools
+    (e.g., `ReadFileTool`, `ShellTool`).
+  - **Discovering tools:** It can also discover tools dynamically:
+    - **Command-based discovery:** If `tools.discoveryCommand` is configured in
+      settings, this command is executed. It's expected to output JSON
+      describing custom tools, which are then registered as `DiscoveredTool`
+      instances.
+    - **MCP-based discovery:** If `mcp.serverCommand` is configured, the
+      registry can connect to a Model Context Protocol (MCP) server to list and
+      register tools (`DiscoveredMCPTool`).
+  - **Providing schemas:** Exposing the `FunctionDeclaration` schemas of all
+    registered tools to the Gemini model, so it knows what tools are available
+    and how to use them.
+  - **Retrieving tools:** Allowing the core to get a specific tool by name for
+    execution.
+
+## Built-in tools
+
+The core comes with a suite of pre-defined tools, typically found in
+`packages/core/src/tools/`. These include:
+
+- **File system tools:**
+  - `LSTool` (`ls.ts`): Lists directory contents.
+  - `ReadFileTool` (`read-file.ts`): Reads the content of a single file.
+  - `WriteFileTool` (`write-file.ts`): Writes content to a file.
+  - `GrepTool` (`grep.ts`): Searches for patterns in files.
+  - `GlobTool` (`glob.ts`): Finds files matching glob patterns.
+  - `EditTool` (`edit.ts`): Performs in-place modifications to files (often
+    requiring confirmation).
+  - `ReadManyFilesTool` (`read-many-files.ts`): Reads and concatenates content
+    from multiple files or glob patterns (used by the `@` command in CLI).
+- **Execution tools:**
+  - `ShellTool` (`shell.ts`): Executes arbitrary shell commands (requires
+    careful sandboxing and user confirmation).
+- **Web tools:**
+  - `WebFetchTool` (`web-fetch.ts`): Fetches content from a URL.
+  - `WebSearchTool` (`web-search.ts`): Performs a web search.
+- **Memory tools:**
+  - `MemoryTool` (`memoryTool.ts`): Interacts with the AI's memory.
+
+Each of these tools extends `BaseTool` and implements the required methods for
+its specific functionality.
+
+## Tool execution flow
+
+1.  **Model request:** The Gemini model, based on the user's prompt and the
+    provided tool schemas, decides to use a tool and returns a `FunctionCall`
+    part in its response, specifying the tool name and arguments.
+2.  **Core receives request:** The core parses this `FunctionCall`.
+3.  **Tool retrieval:** It looks up the requested tool in the `ToolRegistry`.
+4.  **Parameter validation:** The tool's `validateToolParams()` method is
+    called.
+5.  **Confirmation (if needed):**
+    - The tool's `shouldConfirmExecute()` method is called.
+    - If it returns details for confirmation, the core communicates this back to
+      the CLI, which prompts the user.
+    - The user's decision (e.g., proceed, cancel) is sent back to the core.
+6.  **Execution:** If validated and confirmed (or if no confirmation is needed),
+    the core calls the tool's `execute()` method with the provided arguments and
+    an `AbortSignal` (for potential cancellation).
+7.  **Result processing:** The `ToolResult` from `execute()` is received by the
+    core.
+8.  **Response to model:** The `llmContent` from the `ToolResult` is packaged as
+    a `FunctionResponse` and sent back to the Gemini model so it can continue
+    generating a user-facing response.
+9.  **Display to user:** The `returnDisplay` from the `ToolResult` is sent to
+    the CLI to show the user what the tool did.
+
+## Extending with custom tools
+
+While direct programmatic registration of new tools by users isn't explicitly
+detailed as a primary workflow in the provided files for typical end-users, the
+architecture supports extension through:
+
+- **Command-based discovery:** Advanced users or project administrators can
+  define a `tools.discoveryCommand` in `settings.json`. This command, when run
+  by the Gemini CLI core, should output a JSON array of `FunctionDeclaration`
+  objects. The core will then make these available as `DiscoveredTool`
+  instances. The corresponding `tools.callCommand` would then be responsible for
+  actually executing these custom tools.
+- **MCP server(s):** For more complex scenarios, one or more MCP servers can be
+  set up and configured via the `mcpServers` setting in `settings.json`. The
+  Gemini CLI core can then discover and use tools exposed by these servers. As
+  mentioned, if you have multiple MCP servers, the tool names will be prefixed
+  with the server name from your configuration (e.g.,
+  `serverAlias__actualToolName`).
+
+This tool system provides a flexible and powerful way to augment the Gemini
+model's capabilities, making the Gemini CLI a versatile assistant for a wide
+range of tasks.

docs/reference/tools.md

@@ -1,106 +0,0 @@
-# Tools reference
-
-Gemini CLI uses tools to interact with your local environment, access
-information, and perform actions on your behalf. These tools extend the model's
-capabilities beyond text generation, letting it read files, execute commands,
-and search the web.
-
-## How to use Gemini CLI's tools
-
-Tools are generally invoked automatically by Gemini CLI when it needs to perform
-an action. However, you can also trigger specific tools manually using shorthand
-syntax.
-
-### Automatic execution and security
-
-When the model wants to use a tool, Gemini CLI evaluates the request against its
-security policies.
-
-- **User confirmation:** You must manually approve tools that modify files or
-  execute shell commands (mutators). The CLI shows you a diff or the exact
-  command before you confirm.
-- **Sandboxing:** You can run tool executions in secure, containerized
-  environments to isolate changes from your host system. For more details, see
-  the [Sandboxing](../cli/sandbox.md) guide.
-- **Trusted folders:** You can configure which directories allow the model to
-  use system tools. For more details, see the
-  [Trusted folders](../cli/trusted-folders.md) guide.
-
-Review confirmation prompts carefully before allowing a tool to execute.
-
-### How to use manually-triggered tools
-
-You can directly trigger key tools using special syntax in your prompt:
-
-- **[File access](../tools/file-system.md#read_many_files) (`@`):** Use the `@`
-  symbol followed by a file or directory path to include its content in your
-  prompt. This triggers the `read_many_files` tool.
-- **[Shell commands](../tools/shell.md) (`!`):** Use the `!` symbol followed by
-  a system command to execute it directly. This triggers the `run_shell_command`
-  tool.
-
-## How to manage tools
-
-Using built-in commands, you can inspect available tools and configure how they
-behave.
-
-### Tool discovery
-
-Use the `/tools` command to see what tools are currently active in your session.
-
-- **`/tools`**: Lists all registered tools with their display names.
-- **`/tools desc`**: Lists all tools with their full descriptions.
-
-This is especially useful for verifying that
-[MCP servers](../tools/mcp-server.md) or custom tools are loaded correctly.
-
-### Tool configuration
-
-You can enable, disable, or configure specific tools in your settings. For
-example, you can set a specific pager for shell commands or configure the
-browser used for web searches. See the [Settings](../cli/settings.md) guide for
-details.
-
-## Available tools
-
-The following table lists all available tools, categorized by their primary
-function.
-
-| Category    | Tool                                             | Kind          | Description                                                                                                                                                                                                                                 |
-| :---------- | :----------------------------------------------- | :------------ | :------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ |
-| Execution   | [`run_shell_command`](../tools/shell.md)         | `Execute`     | Executes arbitrary shell commands. Supports interactive sessions and background processes. Requires manual confirmation.<br><br>**Parameters:** `command`, `description`, `dir_path`, `is_background`                                       |
-| File System | [`glob`](../tools/file-system.md)                | `Search`      | Finds files matching specific glob patterns across the workspace.<br><br>**Parameters:** `pattern`, `dir_path`, `case_sensitive`, `respect_git_ignore`, `respect_gemini_ignore`                                                             |
-| File System | [`grep_search`](../tools/file-system.md)         | `Search`      | Searches for a regular expression pattern within file contents. Legacy alias: `search_file_content`.<br><br>**Parameters:** `pattern`, `dir_path`, `include`, `exclude_pattern`, `names_only`, `max_matches_per_file`, `total_max_matches`  |
-| File System | [`list_directory`](../tools/file-system.md)      | `Read`        | Lists the names of files and subdirectories within a specified path.<br><br>**Parameters:** `dir_path`, `ignore`, `file_filtering_options`                                                                                                  |
-| File System | [`read_file`](../tools/file-system.md)           | `Read`        | Reads the content of a specific file. Supports text, images, audio, and PDF.<br><br>**Parameters:** `file_path`, `start_line`, `end_line`                                                                                                   |
-| File System | [`read_many_files`](../tools/file-system.md)     | `Read`        | Reads and concatenates content from multiple files. Often triggered by the `@` symbol in your prompt.<br><br>**Parameters:** `include`, `exclude`, `recursive`, `useDefaultExcludes`, `file_filtering_options`                              |
-| File System | [`replace`](../tools/file-system.md)             | `Edit`        | Performs precise text replacement within a file. Requires manual confirmation.<br><br>**Parameters:** `file_path`, `instruction`, `old_string`, `new_string`, `allow_multiple`                                                              |
-| File System | [`write_file`](../tools/file-system.md)          | `Edit`        | Creates or overwrites a file with new content. Requires manual confirmation.<br><br>**Parameters:** `file_path`, `content`                                                                                                                  |
-| Interaction | [`ask_user`](../tools/ask-user.md)               | `Communicate` | Requests clarification or missing information via an interactive dialog.<br><br>**Parameters:** `questions`                                                                                                                                 |
-| Interaction | [`write_todos`](../tools/todos.md)               | `Other`       | Maintains an internal list of subtasks. The model uses this to track its own progress and display it to you.<br><br>**Parameters:** `todos`                                                                                                 |
-| Memory      | [`activate_skill`](../tools/activate-skill.md)   | `Other`       | Loads specialized procedural expertise for specific tasks from the `.gemini/skills` directory.<br><br>**Parameters:** `name`                                                                                                                |
-| Memory      | [`get_internal_docs`](../tools/internal-docs.md) | `Think`       | Accesses Gemini CLI's own documentation to provide more accurate answers about its capabilities.<br><br>**Parameters:** `path`                                                                                                              |
-| Memory      | [`save_memory`](../tools/memory.md)              | `Think`       | Persists specific facts and project details to your `GEMINI.md` file to retain context.<br><br>**Parameters:** `fact`                                                                                                                       |
-| Planning    | [`enter_plan_mode`](../tools/planning.md)        | `Plan`        | Switches the CLI to a safe, read-only "Plan Mode" for researching complex changes.<br><br>**Parameters:** `reason`                                                                                                                          |
-| Planning    | [`exit_plan_mode`](../tools/planning.md)         | `Plan`        | Finalizes a plan, presents it for review, and requests approval to start implementation.<br><br>**Parameters:** `plan`                                                                                                                      |
-| System      | `complete_task`                                  | `Other`       | Finalizes a subagent's mission and returns the result to the parent agent. This tool is not available to the user.<br><br>**Parameters:** `result`                                                                                          |
-| Web         | [`google_web_search`](../tools/web-search.md)    | `Search`      | Performs a Google Search to find up-to-date information.<br><br>**Parameters:** `query`                                                                                                                                                     |
-| Web         | [`web_fetch`](../tools/web-fetch.md)             | `Fetch`       | Retrieves and processes content from specific URLs. **Warning:** This tool can access local and private network addresses (e.g., localhost), which may pose a security risk if used with untrusted prompts.<br><br>**Parameters:** `prompt` |
-
-## Under the hood
-
-For developers, the tool system is designed to be extensible and robust. The
-`ToolRegistry` class manages all available tools.
-
-You can extend Gemini CLI with custom tools by configuring
-`tools.discoveryCommand` in your settings or by connecting to MCP servers.
-
-> **Note:** For a deep dive into the internal Tool API and how to implement your
-> own tools in the codebase, see the `packages/core/src/tools/` directory in
-> GitHub.
-
-## Next steps
-
-- Learn how to [Set up an MCP server](../tools/mcp-server.md).
-- Explore [Agent Skills](../cli/skills.md) for specialized expertise.
-- See the [Command reference](./commands.md) for slash commands.

docs/resources/quota-and-pricing.md

@@ -1,13 +1,14 @@
 # Gemini CLI: Quotas and pricing
 
 Gemini CLI offers a generous free tier that covers many individual developers'
-use cases. For enterprise or professional usage, or if you need increased quota,
+use cases. For enterprise or professional usage, or if you need higher limits,
 several options are available depending on your authentication account type.
 
-For a high-level comparison of available subscriptions and to select the right
-quota for your needs, see the [Plans page](/plans/).
+See [privacy and terms](./tos-privacy.md) for details on the Privacy Policy and
+Terms of Service.
 
-## Overview
+> **Note:** Published prices are list price; additional negotiated commercial
+> discounting may apply.
 
 This article outlines the specific quotas and pricing applicable to Gemini CLI
 when using different authentication methods.
@@ -22,11 +23,10 @@ Generally, there are three categories to choose from:
 
 ## Free usage
 
-Access to Gemini CLI begins with a generous free tier, perfect for
-experimentation and light use.
+Your journey begins with a generous free tier, perfect for experimentation and
+light use.
 
-Your free usage is governed by the following limits, which depend on your
-authorization type.
+Your free usage limits depend on your authorization type.
 
 ### Log in with Google (Gemini Code Assist for individuals)
 
@@ -69,19 +69,6 @@ Learn more at
 If you use up your initial number of requests, you can continue to benefit from
 Gemini CLI by upgrading to one of the following subscriptions:
 
-### Individuals
-
-These tiers apply when you sign in with a personal account. To verify whether
-you're on a personal account, visit
-[Google One](https://one.google.com/about/plans?hl=en-US&g1_landing_page=0):
-
-- If you are on a personal account, you will see your personal dashboard.
-- If you are not on a personal account, you will see: "You're currently signed
-  in to your Google Workspace Account."
-
-**Supported tiers:** _- Tiers not listed above, including Google AI Plus, are
-not supported._
-
 - [Google AI Pro and AI Ultra](https://gemini.google/subscriptions/). This is
   recommended for individual developers. Quotas and pricing are based on a fixed
   price subscription.
@@ -91,26 +78,14 @@ not supported._
   Learn more at
   [Gemini Code Assist Quotas and Limits](https://developers.google.com/gemini-code-assist/resources/quotas)
 
-### Through your organization
-
-These tiers are applicable when you are signing in with a Google Workspace
-account.
-
-- To verify your account type, visit
-  [the Google One page](https://one.google.com/about/plans?hl=en-US&g1_landing_page=0).
-- You are on a workspace account if you see the message "You're currently signed
-  in to your Google Workspace Account".
-
-**Supported tiers:** _- Tiers not listed above, including Workspace AI
-Standard/Plus and AI Expanded, are not supported._
-
-- [Workspace AI Ultra Access](https://workspace.google.com/products/ai-ultra/).
-- [Purchase a Gemini Code Assist Subscription through Google Cloud](https://cloud.google.com/gemini/docs/codeassist/overview).
+- [Purchase a Gemini Code Assist Subscription through Google Cloud ](https://cloud.google.com/gemini/docs/codeassist/overview)
+  by signing up in the Google Cloud console. Learn more at
+  [Set up Gemini Code Assist](https://cloud.google.com/gemini/docs/discover/set-up-gemini).
 
   Quotas and pricing are based on a fixed price subscription with assigned
   license seats. For predictable costs, you can sign in with Google.
 
-  This includes the following request limits:
+  This includes:
   - Gemini Code Assist Standard edition:
     - 1500 model requests / user / day
     - 120 model requests / user / minute
@@ -120,7 +95,7 @@ Standard/Plus and AI Expanded, are not supported._
   - Model requests will be made across the Gemini model family as determined by
     Gemini CLI.
 
-  [Learn more about Gemini Code Assist license limits](https://developers.google.com/gemini-code-assist/resources/quotas#quotas-for-agent-mode-gemini-cli).
+  [Learn more about Gemini Code Assist Standard and Enterprise license limits](https://developers.google.com/gemini-code-assist/resources/quotas#quotas-for-agent-mode-gemini-cli).
 
 ## Pay as you go
 
@@ -131,27 +106,18 @@ recommended path for uninterrupted access.
 
 To do this, log in using a Gemini API key or Vertex AI.
 
-### Vertex AI (regular mode)
-
-An enterprise-grade platform for building, deploying, and managing AI models,
-including Gemini. It offers enhanced security, data governance, and integration
-with other Google Cloud services.
-
-- Quota: Governed by a dynamic shared quota system or pre-purchased provisioned
-  throughput.
-- Cost: Based on model and token usage.
+- Vertex AI (Regular Mode):
+  - Quota: Governed by a dynamic shared quota system or pre-purchased
+    provisioned throughput.
+  - Cost: Based on model and token usage.
 
 Learn more at
 [Vertex AI Dynamic Shared Quota](https://cloud.google.com/vertex-ai/generative-ai/docs/resources/dynamic-shared-quota)
 and [Vertex AI Pricing](https://cloud.google.com/vertex-ai/pricing).
 
-### Gemini API key
-
-Ideal for developers who want to quickly build applications with the Gemini
-models. This is the most direct way to use the models.
-
-- Quota: Varies by pricing tier.
-- Cost: Varies by pricing tier and model/token usage.
+- Gemini API key:
+  - Quota: Varies by pricing tier.
+  - Cost: Varies by pricing tier and model/token usage.
 
 Learn more at
 [Gemini API Rate Limits](https://ai.google.dev/gemini-api/docs/rate-limits),
@@ -159,8 +125,7 @@ Learn more at
 
 It’s important to highlight that when using an API key, you pay per token/call.
 This can be more expensive for many small calls with few tokens, but it's the
-only way to ensure your workflow isn't interrupted by reaching a limit on your
-quota.
+only way to ensure your workflow isn't interrupted by quota limits.
 
 ## Gemini for workspace plans
 
@@ -170,30 +135,31 @@ Flow video editor). These plans do not apply to the API usage which powers the
 Gemini CLI. Supporting these plans is under active consideration for future
 support.
 
-## Check usage and limits
+## Check usage and quota
 
-You can check your current token usage and applicable limits using the
+You can check your current token usage and quota information using the
 `/stats model` command. This command provides a snapshot of your current
-session's token usage, as well as information about the limits associated with
-your current quota.
+session's token usage, as well as your overall quota and usage for the supported
+models.
 
 For more information on the `/stats` command and its subcommands, see the
-[Command Reference](../reference/commands.md#stats).
+[Command Reference](../../reference/commands.md#stats).
 
 A summary of model usage is also presented on exit at the end of a session.
 
 ## Tips to avoid high costs
 
-When using a pay-as-you-go plan, be mindful of your usage to avoid unexpected
+When using a Pay as you Go API key, be mindful of your usage to avoid unexpected
 costs.
 
-- **Be selective with suggestions**: Before accepting a suggestion, especially
-  for a computationally intensive task like refactoring a large codebase,
-  consider if it's the most cost-effective approach.
-- **Use precise prompts**: You are paying per call, so think about the most
-  efficient way to get your desired result. A well-crafted prompt can often get
-  you the answer you need in a single call, rather than multiple back-and-forth
-  interactions.
-- **Monitor your usage**: Use the `/stats model` command to track your token
-  usage during a session. This can help you stay aware of your spending in real
-  time.
+- Don't blindly accept every suggestion, especially for computationally
+  intensive tasks like refactoring large codebases.
+- Be intentional with your prompts and commands. You are paying per call, so
+  think about the most efficient way to get the job done.
+
+## Gemini API vs. Vertex
+
+- Gemini API (gemini developer api): This is the fastest way to use the Gemini
+  models directly.
+- Vertex AI: This is the enterprise-grade platform for building, deploying, and
+  managing Gemini models with specific security and control requirements.

docs/sidebar.json

@@ -106,11 +106,6 @@
           { "label": "MCP servers", "slug": "docs/tools/mcp-server" },
           { "label": "Model routing", "slug": "docs/cli/model-routing" },
           { "label": "Model selection", "slug": "docs/cli/model" },
-          {
-            "label": "Notifications",
-            "badge": "🔬",
-            "slug": "docs/cli/notifications"
-          },
           { "label": "Plan mode", "badge": "🔬", "slug": "docs/cli/plan-mode" },
           {
             "label": "Subagents",
@@ -193,7 +188,7 @@
             "slug": "docs/reference/memport"
           },
           { "label": "Policy engine", "slug": "docs/reference/policy-engine" },
-          { "label": "Tools reference", "slug": "docs/reference/tools" }
+          { "label": "Tools API", "slug": "docs/reference/tools-api" }
         ]
       }
     ]

docs/tools/index.md

@@ -0,0 +1,105 @@
+# Gemini CLI tools
+
+Gemini CLI uses tools to interact with your local environment, access
+information, and perform actions on your behalf. These tools extend the model's
+capabilities beyond text generation, letting it read files, execute commands,
+and search the web.
+
+## User-triggered tools
+
+You can directly trigger these tools using special syntax in your prompts.
+
+- **[File access](./file-system.md#read_many_files) (`@`):** Use the `@` symbol
+  followed by a file or directory path to include its content in your prompt.
+  This triggers the `read_many_files` tool.
+- **[Shell commands](./shell.md) (`!`):** Use the `!` symbol followed by a
+  system command to execute it directly. This triggers the `run_shell_command`
+  tool.
+
+## Model-triggered tools
+
+The Gemini model automatically requests these tools when it needs to perform
+specific actions or gather information to fulfill your requests. You do not call
+these tools manually.
+
+### File management
+
+These tools let the model explore and modify your local codebase.
+
+- **[Directory listing](./file-system.md#list_directory) (`list_directory`):**
+  Lists files and subdirectories.
+- **[File reading](./file-system.md#read_file) (`read_file`):** Reads the
+  content of a specific file.
+- **[File writing](./file-system.md#write_file) (`write_file`):** Creates or
+  overwrites a file with new content.
+- **[File search](./file-system.md#glob) (`glob`):** Finds files matching a glob
+  pattern.
+- **[Text search](./file-system.md#search_file_content)
+  (`search_file_content`):** Searches for text within files using grep or
+  ripgrep.
+- **[Text replacement](./file-system.md#replace) (`replace`):** Performs precise
+  edits within a file.
+
+### Agent coordination
+
+These tools help the model manage its plan and interact with you.
+
+- **Ask user (`ask_user`):** Requests clarification or missing information from
+  you via an interactive dialog.
+- **[Memory](./memory.md) (`save_memory`):** Saves important facts to your
+  long-term memory (`GEMINI.md`).
+- **[Todos](./todos.md) (`write_todos`):** Manages a list of subtasks for
+  complex plans.
+- **[Agent Skills](../cli/skills.md) (`activate_skill`):** Loads specialized
+  procedural expertise when needed.
+- **[Browser agent](../core/subagents.md#browser-agent-experimental)
+  (`browser_agent`):** Automates web browser tasks through the accessibility
+  tree.
+- **Internal docs (`get_internal_docs`):** Accesses Gemini CLI's own
+  documentation to help answer your questions.
+
+### Information gathering
+
+These tools provide the model with access to external data.
+
+- **[Web fetch](./web-fetch.md) (`web_fetch`):** Retrieves and processes content
+  from specific URLs.
+- **[Web search](./web-search.md) (`google_web_search`):** Performs a Google
+  Search to find up-to-date information.
+
+## How to use tools
+
+You use tools indirectly by providing natural language prompts to Gemini CLI.
+
+1.  **Prompt:** You enter a request or use syntax like `@` or `!`.
+2.  **Request:** The model analyzes your request and identifies if a tool is
+    required.
+3.  **Validation:** If a tool is needed, the CLI validates the parameters and
+    checks your security settings.
+4.  **Confirmation:** For sensitive operations (like writing files), the CLI
+    prompts you for approval.
+5.  **Execution:** The tool runs, and its output is sent back to the model.
+6.  **Response:** The model uses the results to generate a final, grounded
+    answer.
+
+## Security and confirmation
+
+Safety is a core part of the tool system. To protect your system, Gemini CLI
+implements several safeguards.
+
+- **User confirmation:** You must manually approve tools that modify files or
+  execute shell commands. The CLI shows you a diff or the exact command before
+  you confirm.
+- **Sandboxing:** You can run tool executions in secure, containerized
+  environments to isolate changes from your host system. For more details, see
+  the [Sandboxing](../cli/sandbox.md) guide.
+- **Trusted folders:** You can configure which directories allow the model to
+  use system tools.
+
+Always review confirmation prompts carefully before allowing a tool to execute.
+
+## Next steps
+
+- Learn how to [Provide context](../cli/gemini-md.md) to guide tool use.
+- Explore the [Command reference](../reference/commands.md) for tool-related
+  slash commands.

docs/tools/mcp-server.md

@@ -372,7 +372,7 @@ To authenticate with a server using Service Account Impersonation, you must set
 the `authProviderType` to `service_account_impersonation` and provide the
 following properties:
 
-- **`targetAudience`** (string): The OAuth Client ID allowlisted on the
+- **`targetAudience`** (string): The OAuth Client ID allowslisted on the
   IAP-protected application you are trying to access.
 - **`targetServiceAccount`** (string): The email address of the Google Cloud
   Service Account to impersonate.

eslint.config.js

@@ -132,16 +132,7 @@ export default tseslint.config(
       'no-cond-assign': 'error',
       'no-debugger': 'error',
       'no-duplicate-case': 'error',
-      'no-restricted-syntax': [
-        'error',
-        ...commonRestrictedSyntaxRules,
-        {
-          selector:
-            'UnaryExpression[operator="typeof"] > MemberExpression[computed=true][property.type="Literal"]',
-          message:
-            'Do not use typeof to check object properties. Define a TypeScript interface and a type guard function instead.',
-        },
-      ],
+      'no-restricted-syntax': ['error', ...commonRestrictedSyntaxRules],
       'no-unsafe-finally': 'error',
       'no-unused-expressions': 'off', // Disable base rule
       '@typescript-eslint/no-unused-expressions': [
@@ -272,7 +263,6 @@ export default tseslint.config(
       ...vitest.configs.recommended.rules,
       'vitest/expect-expect': 'off',
       'vitest/no-commented-out-tests': 'off',
-      'no-restricted-syntax': ['error', ...commonRestrictedSyntaxRules],
     },
   },
   {

evals/concurrency-safety.eval.ts

@@ -1,56 +0,0 @@
-/**
- * @license
- * Copyright 2026 Google LLC
- * SPDX-License-Identifier: Apache-2.0
- */
-
-import { expect } from 'vitest';
-import { evalTest } from './test-helper.js';
-
-const MUTATION_AGENT_DEFINITION = `---
-name: mutation-agent
-description: An agent that modifies the workspace (writes, deletes, git operations, etc).
-max_turns: 1
-tools:
-  - write_file
----
-
-You are the mutation agent. Do the mutation requested.
-`;
-
-describe('concurrency safety eval test cases', () => {
-  evalTest('USUALLY_PASSES', {
-    name: 'mutation agents are run in parallel when explicitly requested',
-    params: {
-      settings: {
-        experimental: {
-          enableAgents: true,
-        },
-      },
-    },
-    prompt:
-      'Update A.txt to say "A" and update B.txt to say "B". Delegate these tasks to two separate mutation-agent subagents. You MUST run these subagents in parallel at the same time.',
-    files: {
-      '.gemini/agents/mutation-agent.md': MUTATION_AGENT_DEFINITION,
-    },
-    assert: async (rig) => {
-      const logs = rig.readToolLogs();
-      const mutationCalls = logs.filter(
-        (log) => log.toolRequest?.name === 'mutation-agent',
-      );
-
-      expect(
-        mutationCalls.length,
-        'Agent should have called the mutation-agent at least twice',
-      ).toBeGreaterThanOrEqual(2);
-
-      const firstPromptId = mutationCalls[0].toolRequest.prompt_id;
-      const secondPromptId = mutationCalls[1].toolRequest.prompt_id;
-
-      expect(
-        firstPromptId,
-        'mutation agents should be called in parallel (same turn / prompt_ids) when explicitly requested',
-      ).toEqual(secondPromptId);
-    },
-  });
-});

evals/test-helper.ts

@@ -112,7 +112,6 @@ export function evalTest(policy: EvalPolicy, evalCase: EvalCase) {
         // commands.
         execSync('git config core.editor "true"', execOptions);
         execSync('git config core.pager "cat"', execOptions);
-        execSync('git config commit.gpgsign false', execOptions);
         execSync('git add .', execOptions);
         execSync('git commit --allow-empty -m "Initial commit"', execOptions);
       }

integration-tests/acp-env-auth.test.ts

@@ -55,7 +55,7 @@ describe.skip('ACP Environment and Auth', () => {
 
       const bundlePath = join(import.meta.dirname, '..', 'bundle/gemini.js');
 
-      child = spawn('node', [bundlePath, '--acp'], {
+      child = spawn('node', [bundlePath, '--experimental-acp'], {
         cwd: rig.homeDir!,
         stdio: ['pipe', 'pipe', 'inherit'],
         env: {
@@ -120,7 +120,7 @@ describe.skip('ACP Environment and Auth', () => {
 
       const bundlePath = join(import.meta.dirname, '..', 'bundle/gemini.js');
 
-      child = spawn('node', [bundlePath, '--acp'], {
+      child = spawn('node', [bundlePath, '--experimental-acp'], {
         cwd: rig.homeDir!,
         stdio: ['pipe', 'pipe', 'inherit'],
         env: {

integration-tests/acp-telemetry.test.ts

@@ -58,7 +58,7 @@ describe('ACP telemetry', () => {
       'node',
       [
         bundlePath,
-        '--acp',
+        '--experimental-acp',
         '--fake-responses',
         join(rig.testDir!, 'fake-responses.json'),
       ],

integration-tests/api-resilience.responses

@@ -1 +0,0 @@
-{"method":"generateContentStream","response":[{"candidates":[{"content":{"parts":[{"text":"Part 1. "}],"role":"model"},"index":0}]},{"usageMetadata":{"promptTokenCount":100,"candidatesTokenCount":10,"totalTokenCount":110}},{"candidates":[{"content":{"parts":[{"text":"Part 2."}],"role":"model"},"index":0}],"finishReason":"STOP"}]}

integration-tests/api-resilience.test.ts

@@ -1,50 +0,0 @@
-/**
- * @license
- * Copyright 2026 Google LLC
- * SPDX-License-Identifier: Apache-2.0
- */
-
-import { describe, it, expect, beforeEach, afterEach } from 'vitest';
-import { TestRig } from './test-helper.js';
-import { join, dirname } from 'node:path';
-import { fileURLToPath } from 'node:url';
-
-describe('API Resilience E2E', () => {
-  let rig: TestRig;
-
-  beforeEach(() => {
-    rig = new TestRig();
-  });
-
-  afterEach(async () => {
-    await rig.cleanup();
-  });
-
-  it('should not crash when receiving metadata-only chunks in a stream', async () => {
-    await rig.setup('api-resilience-metadata-only', {
-      fakeResponsesPath: join(
-        dirname(fileURLToPath(import.meta.url)),
-        'api-resilience.responses',
-      ),
-      settings: {
-        planSettings: { modelRouting: false },
-      },
-    });
-
-    // Run the CLI with a simple prompt.
-    // The fake responses will provide a stream with a metadata-only chunk in the middle.
-    // We use gemini-3-pro-preview to minimize internal service calls.
-    const result = await rig.run({
-      args: ['hi', '--model', 'gemini-3-pro-preview'],
-    });
-
-    // Verify the output contains text from the normal chunks.
-    // If the CLI crashed on the metadata chunk, rig.run would throw.
-    expect(result).toContain('Part 1.');
-    expect(result).toContain('Part 2.');
-
-    // Verify telemetry event for the prompt was still generated
-    const hasUserPromptEvent = await rig.waitForTelemetryEvent('user_prompt');
-    expect(hasUserPromptEvent).toBe(true);
-  });
-});

integration-tests/browser-agent.cleanup.responses

@@ -1,2 +0,0 @@
-{"method":"generateContentStream","response":[{"candidates":[{"content":{"parts":[{"text":"I'll open https://example.com and check the page title for you."},{"functionCall":{"name":"browser_agent","args":{"task":"Open https://example.com and get the page title"}}}],"role":"model"},"finishReason":"STOP","index":0}],"usageMetadata":{"promptTokenCount":100,"candidatesTokenCount":35,"totalTokenCount":135}}]}
-{"method":"generateContentStream","response":[{"candidates":[{"content":{"parts":[{"text":"The page title of https://example.com is \"Example Domain\". The browser session has been completed and cleaned up successfully."}],"role":"model"},"finishReason":"STOP","index":0}],"usageMetadata":{"promptTokenCount":200,"candidatesTokenCount":30,"totalTokenCount":230}}]}

integration-tests/browser-agent.interaction.responses

@@ -1,2 +0,0 @@
-{"method":"generateContentStream","response":[{"candidates":[{"content":{"parts":[{"text":"I'll navigate to https://example.com and analyze the links on the page."},{"functionCall":{"name":"browser_agent","args":{"task":"Go to https://example.com and find all links on the page, then describe them"}}}],"role":"model"},"finishReason":"STOP","index":0}],"usageMetadata":{"promptTokenCount":100,"candidatesTokenCount":40,"totalTokenCount":140}}]}
-{"method":"generateContentStream","response":[{"candidates":[{"content":{"parts":[{"text":"After analyzing https://example.com, I found the following links:\n\n1. **\"More information...\"** - This is the main link on the page that points to the IANA (Internet Assigned Numbers Authority) website for more details about reserved domains.\n\nThe page is quite minimal with just this single informational link, which is typical for example domains used in documentation."}],"role":"model"},"finishReason":"STOP","index":0}],"usageMetadata":{"promptTokenCount":200,"candidatesTokenCount":70,"totalTokenCount":270}}]}

integration-tests/browser-agent.navigate-snapshot.responses

@@ -1,2 +0,0 @@
-{"method":"generateContentStream","response":[{"candidates":[{"content":{"parts":[{"text":"I'll help you open https://example.com and analyze the page. Let me use the browser agent to navigate and capture the page information."},{"functionCall":{"name":"browser_agent","args":{"task":"Navigate to https://example.com and capture the accessibility tree to get the page title and main content"}}}],"role":"model"},"finishReason":"STOP","index":0}],"usageMetadata":{"promptTokenCount":100,"candidatesTokenCount":50,"totalTokenCount":150}}]}
-{"method":"generateContentStream","response":[{"candidates":[{"content":{"parts":[{"text":"Based on the browser analysis of https://example.com:\n\n**Page Title**: Example Domain\n\n**Main Content**: The page contains a simple heading \"Example Domain\" and explanatory text stating \"This domain is for use in illustrative examples in documents. You may use this domain in literature without prior coordination or asking for permission.\"\n\nThe page has a clean, minimal layout typical of placeholder domains used for documentation and examples."}],"role":"model"},"finishReason":"STOP","index":0}],"usageMetadata":{"promptTokenCount":200,"candidatesTokenCount":80,"totalTokenCount":280}}]}

integration-tests/browser-agent.screenshot.responses

@@ -1,2 +0,0 @@
-{"method":"generateContentStream","response":[{"candidates":[{"content":{"parts":[{"text":"I'll navigate to https://example.com and take a screenshot for you."},{"functionCall":{"name":"browser_agent","args":{"task":"Navigate to https://example.com and take a screenshot of the page"}}}],"role":"model"},"finishReason":"STOP","index":0}],"usageMetadata":{"promptTokenCount":100,"candidatesTokenCount":40,"totalTokenCount":140}}]}
-{"method":"generateContentStream","response":[{"candidates":[{"content":{"parts":[{"text":"I've successfully navigated to https://example.com and captured a screenshot. The page shows the familiar \"Example Domain\" header with explanatory text below it. The screenshot captures the clean, minimal layout of this demonstration website."}],"role":"model"},"finishReason":"STOP","index":0}],"usageMetadata":{"promptTokenCount":200,"candidatesTokenCount":50,"totalTokenCount":250}}]}

integration-tests/browser-agent.sequential.responses

@@ -1,2 +0,0 @@
-{"method":"generateContentStream","response":[{"candidates":[{"content":{"parts":[{"text":"I'll help you navigate to https://example.com and perform both operations - capturing the accessibility tree and taking a screenshot."},{"functionCall":{"name":"browser_agent","args":{"task":"Navigate to https://example.com, take a snapshot of the accessibility tree, then take a screenshot"}}}],"role":"model"},"finishReason":"STOP","index":0}],"usageMetadata":{"promptTokenCount":100,"candidatesTokenCount":45,"totalTokenCount":145}}]}
-{"method":"generateContentStream","response":[{"candidates":[{"content":{"parts":[{"text":"I've successfully completed both operations on https://example.com:\n\n**Accessibility Tree Snapshot**: The page has a clear structure with the main heading \"Example Domain\" and descriptive text about the domain's purpose for documentation examples.\n\n**Screenshot**: Captured a visual representation of the page showing the clean, minimal layout with the heading and explanatory text.\n\nBoth the accessibility data and visual screenshot confirm this is the standard example domain page used for documentation purposes."}],"role":"model"},"finishReason":"STOP","index":0}],"usageMetadata":{"promptTokenCount":200,"candidatesTokenCount":80,"totalTokenCount":280}}]}

integration-tests/browser-agent.test.ts

@@ -1,206 +0,0 @@
-/**
- * @license
- * Copyright 2026 Google LLC
- * SPDX-License-Identifier: Apache-2.0
- */
-
-/**
- * Integration tests for the browser agent.
- *
- * These tests verify the complete end-to-end flow from CLI prompt through
- * browser_agent delegation to MCP/Chrome DevTools and back. Unlike the unit
- * tests in packages/core/src/agents/browser/ which mock all MCP components,
- * these tests launch real Chrome instances in headless mode.
- *
- * Tests are skipped on systems without Chrome/Chromium installed.
- */
-
-import { describe, it, expect, beforeEach, afterEach } from 'vitest';
-import { TestRig, assertModelHasOutput } from './test-helper.js';
-import { dirname, join } from 'node:path';
-import { fileURLToPath } from 'node:url';
-import { execSync } from 'node:child_process';
-import { existsSync } from 'node:fs';
-
-const __filename = fileURLToPath(import.meta.url);
-const __dirname = dirname(__filename);
-
-const chromeAvailable = (() => {
-  try {
-    if (process.platform === 'darwin') {
-      execSync(
-        'test -d "/Applications/Google Chrome.app"  || test -d "/Applications/Chromium.app"',
-        {
-          stdio: 'ignore',
-        },
-      );
-    } else if (process.platform === 'linux') {
-      execSync(
-        'which google-chrome || which chromium-browser || which chromium',
-        { stdio: 'ignore' },
-      );
-    } else if (process.platform === 'win32') {
-      // Check standard Windows installation paths using Node.js fs
-      const chromePaths = [
-        'C:\\Program Files\\Google\\Chrome\\Application\\chrome.exe',
-        'C:\\Program Files (x86)\\Google\\Chrome\\Application\\chrome.exe',
-        `${process.env['LOCALAPPDATA'] ?? ''}\\Google\\Chrome\\Application\\chrome.exe`,
-      ];
-      const found = chromePaths.some((p) => existsSync(p));
-      if (!found) {
-        // Fall back to PATH check
-        execSync('where chrome || where chromium', { stdio: 'ignore' });
-      }
-    } else {
-      return false;
-    }
-    return true;
-  } catch {
-    return false;
-  }
-})();
-
-describe.skipIf(!chromeAvailable)('browser-agent', () => {
-  let rig: TestRig;
-
-  beforeEach(() => {
-    rig = new TestRig();
-  });
-
-  afterEach(async () => await rig.cleanup());
-
-  it('should navigate to a page and capture accessibility tree', async () => {
-    rig.setup('browser-navigate-and-snapshot', {
-      fakeResponsesPath: join(
-        __dirname,
-        'browser-agent.navigate-snapshot.responses',
-      ),
-      settings: {
-        agents: {
-          browser_agent: {
-            headless: true,
-            sessionMode: 'isolated',
-          },
-        },
-      },
-    });
-
-    const result = await rig.run({
-      args: 'Open https://example.com in the browser and tell me the page title and main content.',
-    });
-
-    assertModelHasOutput(result);
-
-    const toolLogs = rig.readToolLogs();
-    const browserAgentCall = toolLogs.find(
-      (t) => t.toolRequest.name === 'browser_agent',
-    );
-    expect(
-      browserAgentCall,
-      'Expected browser_agent to be called',
-    ).toBeDefined();
-  });
-
-  it('should take screenshots of web pages', async () => {
-    rig.setup('browser-screenshot', {
-      fakeResponsesPath: join(__dirname, 'browser-agent.screenshot.responses'),
-      settings: {
-        agents: {
-          browser_agent: {
-            headless: true,
-            sessionMode: 'isolated',
-          },
-        },
-      },
-    });
-
-    const result = await rig.run({
-      args: 'Navigate to https://example.com and take a screenshot.',
-    });
-
-    const toolLogs = rig.readToolLogs();
-    const browserCalls = toolLogs.filter(
-      (t) => t.toolRequest.name === 'browser_agent',
-    );
-    expect(browserCalls.length).toBeGreaterThan(0);
-
-    assertModelHasOutput(result);
-  });
-
-  it('should interact with page elements', async () => {
-    rig.setup('browser-interaction', {
-      fakeResponsesPath: join(__dirname, 'browser-agent.interaction.responses'),
-      settings: {
-        agents: {
-          browser_agent: {
-            headless: true,
-            sessionMode: 'isolated',
-          },
-        },
-      },
-    });
-
-    const result = await rig.run({
-      args: 'Go to https://example.com, find any links on the page, and describe them.',
-    });
-
-    const toolLogs = rig.readToolLogs();
-    const browserAgentCall = toolLogs.find(
-      (t) => t.toolRequest.name === 'browser_agent',
-    );
-    expect(
-      browserAgentCall,
-      'Expected browser_agent to be called',
-    ).toBeDefined();
-
-    assertModelHasOutput(result);
-  });
-
-  it('should clean up browser processes after completion', async () => {
-    rig.setup('browser-cleanup', {
-      fakeResponsesPath: join(__dirname, 'browser-agent.cleanup.responses'),
-      settings: {
-        agents: {
-          browser_agent: {
-            headless: true,
-            sessionMode: 'isolated',
-          },
-        },
-      },
-    });
-
-    await rig.run({
-      args: 'Open https://example.com in the browser and check the page title.',
-    });
-
-    // Test passes if we reach here, relying on Vitest's timeout mechanism
-    // to detect hanging browser processes.
-  });
-
-  it('should handle multiple browser operations in sequence', async () => {
-    rig.setup('browser-sequential', {
-      fakeResponsesPath: join(__dirname, 'browser-agent.sequential.responses'),
-      settings: {
-        agents: {
-          browser_agent: {
-            headless: true,
-            sessionMode: 'isolated',
-          },
-        },
-      },
-    });
-
-    const result = await rig.run({
-      args: 'Navigate to https://example.com, take a snapshot of the accessibility tree, then take a screenshot.',
-    });
-
-    const toolLogs = rig.readToolLogs();
-    const browserCalls = toolLogs.filter(
-      (t) => t.toolRequest.name === 'browser_agent',
-    );
-    expect(browserCalls.length).toBeGreaterThan(0);
-
-    // Should successfully complete all operations
-    assertModelHasOutput(result);
-  });
-});

integration-tests/extensions-reload.test.ts

@@ -104,7 +104,7 @@ describe('extension reloading', () => {
           return (
             output.includes(
               'test-server (from test-extension) - Ready (1 tool)',
-            ) && output.includes('- mcp_test-server_hello')
+            ) && output.includes('- hello')
           );
         },
         30000, // 30s timeout
@@ -148,7 +148,7 @@ describe('extension reloading', () => {
           return (
             output.includes(
               'test-server (from test-extension) - Ready (1 tool)',
-            ) && output.includes('- mcp_test-server_goodbye')
+            ) && output.includes('- goodbye')
           );
         },
         30000,

integration-tests/hooks-agent-flow.test.ts

@@ -182,22 +182,14 @@ describe('Hooks Agent Flow', () => {
       );
 
       const afterAgentScript = `
-        const fs = require('fs');
-        const input = JSON.parse(fs.readFileSync(0, 'utf-8'));
-        if (input.stop_hook_active) {
-          // Retry turn: allow execution to proceed (breaks the loop)
-          console.log(JSON.stringify({ decision: 'allow' }));
-        } else {
-          // First call: block and clear context to trigger the retry
-          console.log(JSON.stringify({
-            decision: 'block',
-            reason: 'Security policy triggered',
-            hookSpecificOutput: {
-              hookEventName: 'AfterAgent',
-              clearContext: true
-            }
-          }));
-        }
+        console.log(JSON.stringify({
+          decision: 'block',
+          reason: 'Security policy triggered',
+          hookSpecificOutput: {
+            hookEventName: 'AfterAgent',
+            clearContext: true
+          }
+        }));
       `;
       const afterAgentScriptPath = rig.createScript(
         'after_agent_clear.cjs',
@@ -206,10 +198,8 @@ describe('Hooks Agent Flow', () => {
 
       rig.setup('should process clearContext in AfterAgent hook output', {
         settings: {
-          hooksConfig: {
-            enabled: true,
-          },
           hooks: {
+            enabled: true,
             BeforeModel: [
               {
                 hooks: [

integration-tests/hooks-system.after-agent.responses

@@ -1,3 +1,2 @@
 {"method":"generateContentStream","response":[{"candidates":[{"content":{"parts":[{"text":"Hi there!"}],"role":"model"},"finishReason":"STOP","index":0}]}]}
 {"method":"generateContentStream","response":[{"candidates":[{"content":{"parts":[{"text":"Clarification: I am a bot."}],"role":"model"},"finishReason":"STOP","index":0}]}]}
-{"method":"generateContentStream","response":[{"candidates":[{"content":{"parts":[{"text":"Security policy triggered"}],"role":"model"},"finishReason":"STOP","index":0}]}]}

integration-tests/policy-headless-readonly.responses

@@ -1,2 +0,0 @@
-{"method":"generateContentStream","response":[{"candidates":[{"content":{"parts":[{"text":"I will read the content of the file to identify its"}],"role":"model"},"index":0}],"usageMetadata":{"promptTokenCount":7969,"candidatesTokenCount":11,"totalTokenCount":8061,"promptTokensDetails":[{"modality":"TEXT","tokenCount":7969}],"thoughtsTokenCount":81}},{"candidates":[{"content":{"parts":[{"text":" language.\n"}],"role":"model"},"index":0}],"usageMetadata":{"promptTokenCount":7969,"candidatesTokenCount":14,"totalTokenCount":8064,"promptTokensDetails":[{"modality":"TEXT","tokenCount":7969}],"thoughtsTokenCount":81}},{"candidates":[{"content":{"parts":[{"functionCall":{"name":"read_file","args":{"file_path":"test.txt"}},"thoughtSignature":"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"}],"role":"model"},"index":0}],"usageMetadata":{"promptTokenCount":7969,"candidatesTokenCount":64,"totalTokenCount":8114,"promptTokensDetails":[{"modality":"TEXT","tokenCount":7969}],"thoughtsTokenCount":81}},{"candidates":[{"content":{"parts":[{"text":""}],"role":"model"},"finishReason":"STOP","index":0}],"usageMetadata":{"promptTokenCount":7969,"candidatesTokenCount":64,"totalTokenCount":8114,"cachedContentTokenCount":6082,"promptTokensDetails":[{"modality":"TEXT","tokenCount":7969}],"cacheTokensDetails":[{"modality":"TEXT","tokenCount":6082}],"thoughtsTokenCount":81}}]}
-{"method":"generateContentStream","response":[{"candidates":[{"content":{"parts":[{"text":"The language of the file is Latin."}],"role":"model"},"index":0}],"usageMetadata":{"promptTokenCount":8054,"candidatesTokenCount":8,"totalTokenCount":8078,"promptTokensDetails":[{"modality":"TEXT","tokenCount":8054}],"thoughtsTokenCount":16}},{"candidates":[{"content":{"parts":[{"text":"","thoughtSignature":"EnIKcAG+Pvb7vnRBJVz3khx1oArQQqTNvXOXkliNQS7NvYw94dq5m+wGKRmSj3egO3GVp7pacnAtLn9NT1ABKBGpa7MpRhiAe3bbPZfkqOuveeyC19LKQ9fzasCywiYqg5k5qSxfjs5okk+O0NLOvTjN/tg="}],"role":"model"},"finishReason":"STOP","index":0}],"usageMetadata":{"promptTokenCount":8135,"candidatesTokenCount":8,"totalTokenCount":8159,"promptTokensDetails":[{"modality":"TEXT","tokenCount":8135}],"thoughtsTokenCount":16}}]}

integration-tests/policy-headless-shell-allowed.responses

@@ -1,2 +0,0 @@
-{"method":"generateContentStream","response":[{"candidates":[{"content":{"parts":[{"text":"I will run the requested"}],"role":"model"},"index":0}],"usageMetadata":{"promptTokenCount":7949,"candidatesTokenCount":5,"totalTokenCount":8092,"promptTokensDetails":[{"modality":"TEXT","tokenCount":7949}],"thoughtsTokenCount":138}},{"candidates":[{"content":{"parts":[{"text":" shell command to verify the policy configuration.\n"}],"role":"model"},"index":0}],"usageMetadata":{"promptTokenCount":7949,"candidatesTokenCount":14,"totalTokenCount":8101,"promptTokensDetails":[{"modality":"TEXT","tokenCount":7949}],"thoughtsTokenCount":138}},{"candidates":[{"content":{"parts":[{"functionCall":{"name":"run_shell_command","args":{"command":"echo POLICY_TEST_ECHO_COMMAND","description":"Echo the test string to verify policy settings."}},"thoughtSignature":"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"}],"role":"model"},"index":0}],"usageMetadata":{"promptTokenCount":7949,"candidatesTokenCount":54,"totalTokenCount":8141,"promptTokensDetails":[{"modality":"TEXT","tokenCount":7949}],"thoughtsTokenCount":138}},{"candidates":[{"content":{"parts":[{"text":""}],"role":"model"},"finishReason":"STOP","index":0}],"usageMetadata":{"promptTokenCount":7949,"candidatesTokenCount":54,"totalTokenCount":8141,"cachedContentTokenCount":6082,"promptTokensDetails":[{"modality":"TEXT","tokenCount":7949}],"cacheTokensDetails":[{"modality":"TEXT","tokenCount":6082}],"thoughtsTokenCount":138}}]}
-{"method":"generateContentStream","response":[{"candidates":[{"content":{"parts":[{"text":"POLICY_TEST_"}],"role":"model"},"index":0}],"usageMetadata":{"promptTokenCount":8042,"candidatesTokenCount":4,"totalTokenCount":8046,"promptTokensDetails":[{"modality":"TEXT","tokenCount":8042}]}},{"candidates":[{"content":{"parts":[{"text":"ECHO_COMMAND"}],"role":"model"},"index":0}],"usageMetadata":{"promptTokenCount":8042,"candidatesTokenCount":8,"totalTokenCount":8050,"promptTokensDetails":[{"modality":"TEXT","tokenCount":8042}]}},{"candidates":[{"content":{"parts":[{"text":""}],"role":"model"},"finishReason":"STOP","index":0}],"usageMetadata":{"promptTokenCount":8180,"candidatesTokenCount":8,"totalTokenCount":8188,"promptTokensDetails":[{"modality":"TEXT","tokenCount":8180}]}}]}

integration-tests/policy-headless-shell-denied.responses

@@ -1,2 +0,0 @@
-{"method":"generateContentStream","response":[{"candidates":[{"content":{"parts":[{"text":"**Assessing Command Execution**\n\nOkay, I'm currently assessing the feasibility of executing `echo POLICY_TEST_ECHO_COMMAND` using the `run_shell_command` function. Restrictions are being evaluated; the prompt is specifically geared towards a successful command output: \"POLICY_TEST_ECHO_COMMAND\".\n\n\n","thought":true}],"role":"model"},"index":0}],"usageMetadata":{"promptTokenCount":7949,"totalTokenCount":7949,"promptTokensDetails":[{"modality":"TEXT","tokenCount":7949}]}},{"candidates":[{"content":{"parts":[{"text":"I will execute the requested echo"}],"role":"model"},"index":0}],"usageMetadata":{"promptTokenCount":7949,"candidatesTokenCount":6,"totalTokenCount":8161,"promptTokensDetails":[{"modality":"TEXT","tokenCount":7949}],"thoughtsTokenCount":206}},{"candidates":[{"content":{"parts":[{"text":" command to verify the policy."}],"role":"model"},"index":0}],"usageMetadata":{"promptTokenCount":7949,"candidatesTokenCount":12,"totalTokenCount":8167,"promptTokensDetails":[{"modality":"TEXT","tokenCount":7949}],"thoughtsTokenCount":206}},{"candidates":[{"content":{"parts":[{"functionCall":{"name":"run_shell_command","args":{"description":"Execute the echo command as requested.","command":"echo POLICY_TEST_ECHO_COMMAND"}},"thoughtSignature":"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"}],"role":"model"},"index":0}],"usageMetadata":{"promptTokenCount":7949,"candidatesTokenCount":50,"totalTokenCount":8205,"promptTokensDetails":[{"modality":"TEXT","tokenCount":7949}],"thoughtsTokenCount":206}},{"candidates":[{"content":{"parts":[{"text":""}],"role":"model"},"finishReason":"STOP","index":0}],"usageMetadata":{"promptTokenCount":7949,"candidatesTokenCount":50,"totalTokenCount":8205,"cachedContentTokenCount":6082,"promptTokensDetails":[{"modality":"TEXT","tokenCount":7949}],"cacheTokensDetails":[{"modality":"TEXT","tokenCount":6082}],"thoughtsTokenCount":206}}]}
-{"method":"generateContentStream","response":[{"candidates":[{"content":{"parts":[{"text":"AR NAR"}],"role":"model"},"index":0}],"usageMetadata":{"promptTokenCount":8020,"candidatesTokenCount":2,"totalTokenCount":8049,"promptTokensDetails":[{"modality":"TEXT","tokenCount":8020}],"thoughtsTokenCount":27}},{"candidates":[{"content":{"parts":[{"text":"","thoughtSignature":"Er8BCrwBAb4+9vv6KGeMf6yopmPBE/az7Kjdp+Pe5a/R6wgXcyCZzGNwkwKFW3i3ro0j26bRrVeHD1zRfWFTIGdOSZKV6OMPWLqFC/RU6CNJ88B1xY7hbCVwA7EchYPzgd3YZRVNwmFu52j86/9qXf/zaqTFN+WQ0mUESJXh2O2YX8E7imAvxhmRdobVkxvEt4ZX3dW5skDhXHMDZOxbLpX0nkK7cWWS7iEc+qBFP0yinlA/eiG2ZdKpuTiDl76a9ik="}],"role":"model"},"finishReason":"STOP","index":0}],"usageMetadata":{"promptTokenCount":8226,"candidatesTokenCount":2,"totalTokenCount":8255,"promptTokensDetails":[{"modality":"TEXT","tokenCount":8226}],"thoughtsTokenCount":27}}]}

integration-tests/policy-headless.test.ts

@@ -1,205 +0,0 @@
-/**
- * @license
- * Copyright 2026 Google LLC
- * SPDX-License-Identifier: Apache-2.0
- */
-
-import { describe, it, expect, beforeEach, afterEach } from 'vitest';
-import { join } from 'node:path';
-import { TestRig } from './test-helper.js';
-
-interface PromptCommand {
-  prompt: (testFile: string) => string;
-  tool: string;
-  command: string;
-  expectedSuccessResult: string;
-  expectedFailureResult: string;
-}
-
-const ECHO_PROMPT: PromptCommand = {
-  command: 'echo',
-  prompt: () =>
-    `Use the \`echo POLICY_TEST_ECHO_COMMAND\` shell command. On success, ` +
-    `your final response must ONLY be "POLICY_TEST_ECHO_COMMAND". If the ` +
-    `command fails output AR NAR and stop.`,
-  tool: 'run_shell_command',
-  expectedSuccessResult: 'POLICY_TEST_ECHO_COMMAND',
-  expectedFailureResult: 'AR NAR',
-};
-
-const READ_FILE_PROMPT: PromptCommand = {
-  prompt: (testFile: string) =>
-    `Read the file ${testFile} and tell me what language it is, if the ` +
-    `read_file tool fails output AR NAR and stop.`,
-  tool: 'read_file',
-  command: '',
-  expectedSuccessResult: 'Latin',
-  expectedFailureResult: 'AR NAR',
-};
-
-async function waitForToolCallLog(
-  rig: TestRig,
-  tool: string,
-  command: string,
-  timeout: number = 15000,
-) {
-  const foundToolCall = await rig.waitForToolCall(tool, timeout, (args) =>
-    args.toLowerCase().includes(command.toLowerCase()),
-  );
-
-  expect(foundToolCall).toBe(true);
-
-  const toolLogs = rig
-    .readToolLogs()
-    .filter((toolLog) => toolLog.toolRequest.name === tool);
-  const log = toolLogs.find(
-    (toolLog) =>
-      !command ||
-      toolLog.toolRequest.args.toLowerCase().includes(command.toLowerCase()),
-  );
-
-  // The policy engine should have logged the tool call
-  expect(log).toBeTruthy();
-  return log;
-}
-
-async function verifyToolExecution(
-  rig: TestRig,
-  promptCommand: PromptCommand,
-  result: string,
-  expectAllowed: boolean,
-  expectedDenialString?: string,
-) {
-  const log = await waitForToolCallLog(
-    rig,
-    promptCommand.tool,
-    promptCommand.command,
-  );
-
-  if (expectAllowed) {
-    expect(log!.toolRequest.success).toBe(true);
-    expect(result).not.toContain('Tool execution denied by policy');
-    expect(result).not.toContain(`Tool "${promptCommand.tool}" not found`);
-    expect(result).toContain(promptCommand.expectedSuccessResult);
-  } else {
-    expect(log!.toolRequest.success).toBe(false);
-    expect(result).toContain(
-      expectedDenialString || 'Tool execution denied by policy',
-    );
-    expect(result).toContain(promptCommand.expectedFailureResult);
-  }
-}
-
-interface TestCase {
-  name: string;
-  responsesFile: string;
-  promptCommand: PromptCommand;
-  policyContent?: string;
-  expectAllowed: boolean;
-  expectedDenialString?: string;
-}
-
-describe('Policy Engine Headless Mode', () => {
-  let rig: TestRig;
-  let testFile: string;
-
-  beforeEach(() => {
-    rig = new TestRig();
-  });
-
-  afterEach(async () => {
-    if (rig) {
-      await rig.cleanup();
-    }
-  });
-
-  const runTestCase = async (tc: TestCase) => {
-    const fakeResponsesPath = join(import.meta.dirname, tc.responsesFile);
-    rig.setup(tc.name, { fakeResponsesPath });
-
-    testFile = rig.createFile('test.txt', 'Lorem\nIpsum\nDolor\n');
-    const args = ['-p', tc.promptCommand.prompt(testFile)];
-
-    if (tc.policyContent) {
-      const policyPath = rig.createFile('test-policy.toml', tc.policyContent);
-      args.push('--policy', policyPath);
-    }
-
-    const result = await rig.run({
-      args,
-      approvalMode: 'default',
-    });
-
-    await verifyToolExecution(
-      rig,
-      tc.promptCommand,
-      result,
-      tc.expectAllowed,
-      tc.expectedDenialString,
-    );
-  };
-
-  const testCases = [
-    {
-      name: 'should deny ASK_USER tools by default in headless mode',
-      responsesFile: 'policy-headless-shell-denied.responses',
-      promptCommand: ECHO_PROMPT,
-      expectAllowed: false,
-      expectedDenialString: 'Tool "run_shell_command" not found',
-    },
-    {
-      name: 'should allow ASK_USER tools in headless mode if explicitly allowed via policy file',
-      responsesFile: 'policy-headless-shell-allowed.responses',
-      promptCommand: ECHO_PROMPT,
-      policyContent: `
-      [[rule]]
-      toolName = "run_shell_command"
-      decision = "allow"
-      priority = 100
-    `,
-      expectAllowed: true,
-    },
-    {
-      name: 'should allow read-only tools by default in headless mode',
-      responsesFile: 'policy-headless-readonly.responses',
-      promptCommand: READ_FILE_PROMPT,
-      expectAllowed: true,
-    },
-    {
-      name: 'should allow specific shell commands in policy file',
-      responsesFile: 'policy-headless-shell-allowed.responses',
-      promptCommand: ECHO_PROMPT,
-      policyContent: `
-        [[rule]]
-        toolName = "run_shell_command"
-        commandPrefix = "${ECHO_PROMPT.command}"
-        decision = "allow"
-        priority = 100
-      `,
-      expectAllowed: true,
-    },
-    {
-      name: 'should deny other shell commands in policy file',
-      responsesFile: 'policy-headless-shell-denied.responses',
-      promptCommand: ECHO_PROMPT,
-      policyContent: `
-        [[rule]]
-        toolName = "run_shell_command"
-        commandPrefix = "node"
-        decision = "allow"
-        priority = 100
-      `,
-      expectAllowed: false,
-      expectedDenialString: 'Tool execution denied by policy',
-    },
-  ];
-
-  it.each(testCases)(
-    '$name',
-    async (tc) => {
-      await runTestCase(tc);
-    },
-    // Large timeout for regeneration
-    process.env['REGENERATE_MODEL_GOLDENS'] === 'true' ? 120000 : undefined,
-  );
-});

packages/a2a-server/src/agent/task.ts

@@ -28,9 +28,6 @@ import {
   type Config,
   type UserTierId,
   type ToolLiveOutput,
-  type AnsiLine,
-  type AnsiOutput,
-  type AnsiToken,
   isSubagentProgress,
   EDIT_TOOL_NAMES,
   processRestorableToolCalls,
@@ -347,15 +344,10 @@ export class Task {
       outputAsText = outputChunk;
     } else if (isSubagentProgress(outputChunk)) {
       outputAsText = JSON.stringify(outputChunk);
-    } else if (Array.isArray(outputChunk)) {
-      const ansiOutput: AnsiOutput = outputChunk;
-      outputAsText = ansiOutput
-        .map((line: AnsiLine) =>
-          line.map((token: AnsiToken) => token.text).join(''),
-        )
-        .join('\n');
     } else {
-      outputAsText = String(outputChunk);
+      outputAsText = outputChunk
+        .map((line) => line.map((token) => token.text).join(''))
+        .join('\n');
     }
 
     logger.info(
@@ -832,9 +824,7 @@ export class Task {
     if (
       part.kind !== 'data' ||
       !part.data ||
-      // eslint-disable-next-line no-restricted-syntax
       typeof part.data['callId'] !== 'string' ||
-      // eslint-disable-next-line no-restricted-syntax
       typeof part.data['outcome'] !== 'string'
     ) {
       return false;

packages/a2a-server/src/config/config.test.ts

@@ -16,9 +16,6 @@ import {
   ExperimentFlags,
   fetchAdminControlsOnce,
   type FetchAdminControlsResponse,
-  AuthType,
-  isHeadlessMode,
-  FatalAuthenticationError,
 } from '@google/gemini-cli-core';
 
 // Mock dependencies
@@ -53,7 +50,6 @@ vi.mock('@google/gemini-cli-core', async (importOriginal) => {
     startupProfiler: {
       flush: vi.fn(),
     },
-    isHeadlessMode: vi.fn().mockReturnValue(false),
     FileDiscoveryService: vi.fn(),
     getCodeAssistServer: vi.fn(),
     fetchAdminControlsOnce: vi.fn(),
@@ -66,7 +62,6 @@ vi.mock('@google/gemini-cli-core', async (importOriginal) => {
 vi.mock('../utils/logger.js', () => ({
   logger: {
     info: vi.fn(),
-    warn: vi.fn(),
     error: vi.fn(),
   },
 }));
@@ -78,11 +73,12 @@ describe('loadConfig', () => {
 
   beforeEach(() => {
     vi.clearAllMocks();
-    vi.stubEnv('GEMINI_API_KEY', 'test-key');
+    process.env['GEMINI_API_KEY'] = 'test-key';
   });
 
   afterEach(() => {
-    vi.unstubAllEnvs();
+    delete process.env['CUSTOM_IGNORE_FILE_PATHS'];
+    delete process.env['GEMINI_API_KEY'];
   });
 
   describe('admin settings overrides', () => {
@@ -203,7 +199,7 @@ describe('loadConfig', () => {
 
   it('should set customIgnoreFilePaths when CUSTOM_IGNORE_FILE_PATHS env var is present', async () => {
     const testPath = '/tmp/ignore';
-    vi.stubEnv('CUSTOM_IGNORE_FILE_PATHS', testPath);
+    process.env['CUSTOM_IGNORE_FILE_PATHS'] = testPath;
     const config = await loadConfig(mockSettings, mockExtensionLoader, taskId);
     // eslint-disable-next-line @typescript-eslint/no-explicit-any
     expect((config as any).fileFiltering.customIgnoreFilePaths).toEqual([
@@ -228,7 +224,7 @@ describe('loadConfig', () => {
   it('should merge customIgnoreFilePaths from settings and env var', async () => {
     const envPath = '/env/ignore';
     const settingsPath = '/settings/ignore';
-    vi.stubEnv('CUSTOM_IGNORE_FILE_PATHS', envPath);
+    process.env['CUSTOM_IGNORE_FILE_PATHS'] = envPath;
     const settings: Settings = {
       fileFiltering: {
         customIgnoreFilePaths: [settingsPath],
@@ -244,7 +240,7 @@ describe('loadConfig', () => {
 
   it('should split CUSTOM_IGNORE_FILE_PATHS using system delimiter', async () => {
     const paths = ['/path/one', '/path/two'];
-    vi.stubEnv('CUSTOM_IGNORE_FILE_PATHS', paths.join(path.delimiter));
+    process.env['CUSTOM_IGNORE_FILE_PATHS'] = paths.join(path.delimiter);
     const config = await loadConfig(mockSettings, mockExtensionLoader, taskId);
     // eslint-disable-next-line @typescript-eslint/no-explicit-any
     expect((config as any).fileFiltering.customIgnoreFilePaths).toEqual(paths);
@@ -258,7 +254,7 @@ describe('loadConfig', () => {
 
   it('should initialize FileDiscoveryService with correct options', async () => {
     const testPath = '/tmp/ignore';
-    vi.stubEnv('CUSTOM_IGNORE_FILE_PATHS', testPath);
+    process.env['CUSTOM_IGNORE_FILE_PATHS'] = testPath;
     const settings: Settings = {
       fileFiltering: {
         respectGitIgnore: false,
@@ -315,219 +311,5 @@ describe('loadConfig', () => {
         }),
       );
     });
-
-    describe('interactivity', () => {
-      it('should set interactive true when not headless', async () => {
-        vi.mocked(isHeadlessMode).mockReturnValue(false);
-        await loadConfig(mockSettings, mockExtensionLoader, taskId);
-        expect(Config).toHaveBeenCalledWith(
-          expect.objectContaining({
-            interactive: true,
-            enableInteractiveShell: true,
-          }),
-        );
-      });
-
-      it('should set interactive false when headless', async () => {
-        vi.mocked(isHeadlessMode).mockReturnValue(true);
-        await loadConfig(mockSettings, mockExtensionLoader, taskId);
-        expect(Config).toHaveBeenCalledWith(
-          expect.objectContaining({
-            interactive: false,
-            enableInteractiveShell: false,
-          }),
-        );
-      });
-    });
-
-    describe('authentication fallback', () => {
-      beforeEach(() => {
-        vi.stubEnv('USE_CCPA', 'true');
-        vi.stubEnv('GEMINI_API_KEY', '');
-      });
-
-      afterEach(() => {
-        vi.unstubAllEnvs();
-      });
-
-      it('should fall back to COMPUTE_ADC in Cloud Shell if LOGIN_WITH_GOOGLE fails', async () => {
-        vi.stubEnv('CLOUD_SHELL', 'true');
-        vi.mocked(isHeadlessMode).mockReturnValue(false);
-        const refreshAuthMock = vi.fn().mockImplementation((authType) => {
-          if (authType === AuthType.LOGIN_WITH_GOOGLE) {
-            throw new FatalAuthenticationError('Non-interactive session');
-          }
-          return Promise.resolve();
-        });
-
-        // Update the mock implementation for this test
-        vi.mocked(Config).mockImplementation(
-          (params: unknown) =>
-            ({
-              ...(params as object),
-              initialize: vi.fn(),
-              waitForMcpInit: vi.fn(),
-              refreshAuth: refreshAuthMock,
-              getExperiments: vi.fn().mockReturnValue({ flags: {} }),
-              getRemoteAdminSettings: vi.fn(),
-              setRemoteAdminSettings: vi.fn(),
-            }) as unknown as Config,
-        );
-
-        await loadConfig(mockSettings, mockExtensionLoader, taskId);
-
-        expect(refreshAuthMock).toHaveBeenCalledWith(
-          AuthType.LOGIN_WITH_GOOGLE,
-        );
-        expect(refreshAuthMock).toHaveBeenCalledWith(AuthType.COMPUTE_ADC);
-      });
-
-      it('should not fall back to COMPUTE_ADC if not in cloud environment', async () => {
-        vi.mocked(isHeadlessMode).mockReturnValue(false);
-        const refreshAuthMock = vi.fn().mockImplementation((authType) => {
-          if (authType === AuthType.LOGIN_WITH_GOOGLE) {
-            throw new FatalAuthenticationError('Non-interactive session');
-          }
-          return Promise.resolve();
-        });
-
-        vi.mocked(Config).mockImplementation(
-          (params: unknown) =>
-            ({
-              ...(params as object),
-              initialize: vi.fn(),
-              waitForMcpInit: vi.fn(),
-              refreshAuth: refreshAuthMock,
-              getExperiments: vi.fn().mockReturnValue({ flags: {} }),
-              getRemoteAdminSettings: vi.fn(),
-              setRemoteAdminSettings: vi.fn(),
-            }) as unknown as Config,
-        );
-
-        await expect(
-          loadConfig(mockSettings, mockExtensionLoader, taskId),
-        ).rejects.toThrow('Non-interactive session');
-
-        expect(refreshAuthMock).toHaveBeenCalledWith(
-          AuthType.LOGIN_WITH_GOOGLE,
-        );
-        expect(refreshAuthMock).not.toHaveBeenCalledWith(AuthType.COMPUTE_ADC);
-      });
-
-      it('should skip LOGIN_WITH_GOOGLE and use COMPUTE_ADC directly in headless Cloud Shell', async () => {
-        vi.stubEnv('CLOUD_SHELL', 'true');
-        vi.mocked(isHeadlessMode).mockReturnValue(true);
-
-        const refreshAuthMock = vi.fn().mockResolvedValue(undefined);
-
-        vi.mocked(Config).mockImplementation(
-          (params: unknown) =>
-            ({
-              ...(params as object),
-              initialize: vi.fn(),
-              waitForMcpInit: vi.fn(),
-              refreshAuth: refreshAuthMock,
-              getExperiments: vi.fn().mockReturnValue({ flags: {} }),
-              getRemoteAdminSettings: vi.fn(),
-              setRemoteAdminSettings: vi.fn(),
-            }) as unknown as Config,
-        );
-
-        await loadConfig(mockSettings, mockExtensionLoader, taskId);
-
-        expect(refreshAuthMock).not.toHaveBeenCalledWith(
-          AuthType.LOGIN_WITH_GOOGLE,
-        );
-        expect(refreshAuthMock).toHaveBeenCalledWith(AuthType.COMPUTE_ADC);
-      });
-
-      it('should skip LOGIN_WITH_GOOGLE and use COMPUTE_ADC directly if GEMINI_CLI_USE_COMPUTE_ADC is true', async () => {
-        vi.stubEnv('GEMINI_CLI_USE_COMPUTE_ADC', 'true');
-        vi.mocked(isHeadlessMode).mockReturnValue(false); // Even if not headless
-
-        const refreshAuthMock = vi.fn().mockResolvedValue(undefined);
-
-        vi.mocked(Config).mockImplementation(
-          (params: unknown) =>
-            ({
-              ...(params as object),
-              initialize: vi.fn(),
-              waitForMcpInit: vi.fn(),
-              refreshAuth: refreshAuthMock,
-              getExperiments: vi.fn().mockReturnValue({ flags: {} }),
-              getRemoteAdminSettings: vi.fn(),
-              setRemoteAdminSettings: vi.fn(),
-            }) as unknown as Config,
-        );
-
-        await loadConfig(mockSettings, mockExtensionLoader, taskId);
-
-        expect(refreshAuthMock).not.toHaveBeenCalledWith(
-          AuthType.LOGIN_WITH_GOOGLE,
-        );
-        expect(refreshAuthMock).toHaveBeenCalledWith(AuthType.COMPUTE_ADC);
-      });
-
-      it('should throw FatalAuthenticationError in headless mode if no ADC fallback available', async () => {
-        vi.mocked(isHeadlessMode).mockReturnValue(true);
-
-        const refreshAuthMock = vi.fn().mockResolvedValue(undefined);
-
-        vi.mocked(Config).mockImplementation(
-          (params: unknown) =>
-            ({
-              ...(params as object),
-              initialize: vi.fn(),
-              waitForMcpInit: vi.fn(),
-              refreshAuth: refreshAuthMock,
-              getExperiments: vi.fn().mockReturnValue({ flags: {} }),
-              getRemoteAdminSettings: vi.fn(),
-              setRemoteAdminSettings: vi.fn(),
-            }) as unknown as Config,
-        );
-
-        await expect(
-          loadConfig(mockSettings, mockExtensionLoader, taskId),
-        ).rejects.toThrow(
-          'Interactive terminal required for LOGIN_WITH_GOOGLE. Run in an interactive terminal or set GEMINI_CLI_USE_COMPUTE_ADC=true to use Application Default Credentials.',
-        );
-
-        expect(refreshAuthMock).not.toHaveBeenCalled();
-      });
-
-      it('should include both original and fallback error when COMPUTE_ADC fallback fails', async () => {
-        vi.stubEnv('CLOUD_SHELL', 'true');
-        vi.mocked(isHeadlessMode).mockReturnValue(false);
-
-        const refreshAuthMock = vi.fn().mockImplementation((authType) => {
-          if (authType === AuthType.LOGIN_WITH_GOOGLE) {
-            throw new FatalAuthenticationError('OAuth failed');
-          }
-          if (authType === AuthType.COMPUTE_ADC) {
-            throw new Error('ADC failed');
-          }
-          return Promise.resolve();
-        });
-
-        vi.mocked(Config).mockImplementation(
-          (params: unknown) =>
-            ({
-              ...(params as object),
-              initialize: vi.fn(),
-              waitForMcpInit: vi.fn(),
-              refreshAuth: refreshAuthMock,
-              getExperiments: vi.fn().mockReturnValue({ flags: {} }),
-              getRemoteAdminSettings: vi.fn(),
-              setRemoteAdminSettings: vi.fn(),
-            }) as unknown as Config,
-        );
-
-        await expect(
-          loadConfig(mockSettings, mockExtensionLoader, taskId),
-        ).rejects.toThrow(
-          'OAuth failed. Fallback to COMPUTE_ADC also failed: ADC failed',
-        );
-      });
-    });
   });
 });

packages/a2a-server/src/config/config.ts

@@ -23,9 +23,6 @@ import {
   fetchAdminControlsOnce,
   getCodeAssistServer,
   ExperimentFlags,
-  isHeadlessMode,
-  FatalAuthenticationError,
-  isCloudShell,
   type TelemetryTarget,
   type ConfigParameters,
   type ExtensionLoader,
@@ -106,8 +103,8 @@ export async function loadConfig(
     trustedFolder: true,
     extensionLoader,
     checkpointing,
-    interactive: !isHeadlessMode(),
-    enableInteractiveShell: !isHeadlessMode(),
+    interactive: true,
+    enableInteractiveShell: true,
     ptyInfo: 'auto',
   };
 
@@ -120,6 +117,7 @@ export async function loadConfig(
     await loadServerHierarchicalMemory(
       workspaceDir,
       [workspaceDir],
+      false,
       fileService,
       extensionLoader,
       folderTrust,
@@ -257,61 +255,7 @@ async function refreshAuthentication(
         `[${logPrefix}] USE_CCPA env var is true but unable to resolve GOOGLE_APPLICATION_CREDENTIALS file path ${adcFilePath}. Error ${e}`,
       );
     }
-
-    const useComputeAdc = process.env['GEMINI_CLI_USE_COMPUTE_ADC'] === 'true';
-    const isHeadless = isHeadlessMode();
-    const shouldSkipOauth = isHeadless || useComputeAdc;
-
-    if (shouldSkipOauth) {
-      if (isCloudShell() || useComputeAdc) {
-        logger.info(
-          `[${logPrefix}] Skipping LOGIN_WITH_GOOGLE due to ${isHeadless ? 'headless mode' : 'GEMINI_CLI_USE_COMPUTE_ADC'}. Attempting COMPUTE_ADC.`,
-        );
-        try {
-          await config.refreshAuth(AuthType.COMPUTE_ADC);
-          logger.info(`[${logPrefix}] COMPUTE_ADC successful.`);
-        } catch (adcError) {
-          const adcMessage =
-            adcError instanceof Error ? adcError.message : String(adcError);
-          throw new FatalAuthenticationError(
-            `COMPUTE_ADC failed: ${adcMessage}. (Skipped LOGIN_WITH_GOOGLE due to ${isHeadless ? 'headless mode' : 'GEMINI_CLI_USE_COMPUTE_ADC'})`,
-          );
-        }
-      } else {
-        throw new FatalAuthenticationError(
-          `Interactive terminal required for LOGIN_WITH_GOOGLE. Run in an interactive terminal or set GEMINI_CLI_USE_COMPUTE_ADC=true to use Application Default Credentials.`,
-        );
-      }
-    } else {
-      try {
-        await config.refreshAuth(AuthType.LOGIN_WITH_GOOGLE);
-      } catch (e) {
-        if (
-          e instanceof FatalAuthenticationError &&
-          (isCloudShell() || useComputeAdc)
-        ) {
-          logger.warn(
-            `[${logPrefix}] LOGIN_WITH_GOOGLE failed. Attempting COMPUTE_ADC fallback.`,
-          );
-          try {
-            await config.refreshAuth(AuthType.COMPUTE_ADC);
-            logger.info(`[${logPrefix}] COMPUTE_ADC fallback successful.`);
-          } catch (adcError) {
-            logger.error(
-              `[${logPrefix}] COMPUTE_ADC fallback failed: ${adcError}`,
-            );
-            const originalMessage = e instanceof Error ? e.message : String(e);
-            const adcMessage =
-              adcError instanceof Error ? adcError.message : String(adcError);
-            throw new FatalAuthenticationError(
-              `${originalMessage}. Fallback to COMPUTE_ADC also failed: ${adcMessage}`,
-            );
-          }
-        } else {
-          throw e;
-        }
-      }
-    }
+    await config.refreshAuth(AuthType.LOGIN_WITH_GOOGLE);
     logger.info(
       `[${logPrefix}] GOOGLE_CLOUD_PROJECT: ${process.env['GOOGLE_CLOUD_PROJECT']}`,
     );

packages/cli/src/commands/extensions/install.ts

@@ -5,7 +5,6 @@
  */
 
 import type { CommandModule } from 'yargs';
-import * as path from 'node:path';
 import chalk from 'chalk';
 import {
   debugLogger,
@@ -52,13 +51,12 @@ export async function handleInstall(args: InstallArgs) {
     const settings = loadSettings(workspaceDir).merged;
 
     if (installMetadata.type === 'local' || installMetadata.type === 'link') {
-      const absolutePath = path.resolve(source);
-      const realPath = getRealPath(absolutePath);
-      installMetadata.source = absolutePath;
-      const trustResult = isWorkspaceTrusted(settings, absolutePath);
+      const resolvedPath = getRealPath(source);
+      installMetadata.source = resolvedPath;
+      const trustResult = isWorkspaceTrusted(settings, resolvedPath);
       if (trustResult.isTrusted !== true) {
         const discoveryResults =
-          await FolderTrustDiscoveryService.discover(realPath);
+          await FolderTrustDiscoveryService.discover(resolvedPath);
 
         const hasDiscovery =
           discoveryResults.commands.length > 0 ||
@@ -71,7 +69,7 @@ export async function handleInstall(args: InstallArgs) {
           '',
           chalk.bold('Do you trust the files in this folder?'),
           '',
-          `The extension source at "${absolutePath}" is not trusted.`,
+          `The extension source at "${resolvedPath}" is not trusted.`,
           '',
           'Trusting a folder allows Gemini CLI to load its local configurations,',
           'including custom commands, hooks, MCP servers, agent skills, and',
@@ -129,10 +127,10 @@ export async function handleInstall(args: InstallArgs) {
         );
         if (confirmed) {
           const trustedFolders = loadTrustedFolders();
-          await trustedFolders.setValue(realPath, TrustLevel.TRUST_FOLDER);
+          await trustedFolders.setValue(resolvedPath, TrustLevel.TRUST_FOLDER);
         } else {
           throw new Error(
-            `Installation aborted: Folder "${absolutePath}" is not trusted.`,
+            `Installation aborted: Folder "${resolvedPath}" is not trusted.`,
           );
         }
       }

packages/cli/src/commands/extensions/uninstall.test.ts

@@ -28,7 +28,6 @@ import { getErrorMessage } from '../../utils/errors.js';
 // Hoisted mocks - these survive vi.clearAllMocks()
 const mockUninstallExtension = vi.hoisted(() => vi.fn());
 const mockLoadExtensions = vi.hoisted(() => vi.fn());
-const mockGetExtensions = vi.hoisted(() => vi.fn());
 
 // Mock dependencies with hoisted functions
 vi.mock('../../config/extension-manager.js', async (importOriginal) => {
@@ -39,7 +38,6 @@ vi.mock('../../config/extension-manager.js', async (importOriginal) => {
     ExtensionManager: vi.fn().mockImplementation(() => ({
       uninstallExtension: mockUninstallExtension,
       loadExtensions: mockLoadExtensions,
-      getExtensions: mockGetExtensions,
       setRequestConsent: vi.fn(),
       setRequestSetting: vi.fn(),
     })),
@@ -95,7 +93,6 @@ describe('extensions uninstall command', () => {
   afterEach(() => {
     mockLoadExtensions.mockClear();
     mockUninstallExtension.mockClear();
-    mockGetExtensions.mockClear();
     vi.clearAllMocks();
   });
 
@@ -148,41 +145,6 @@ describe('extensions uninstall command', () => {
       mockCwd.mockRestore();
     });
 
-    it('should uninstall all extensions when --all flag is used', async () => {
-      mockLoadExtensions.mockResolvedValue(undefined);
-      mockUninstallExtension.mockResolvedValue(undefined);
-      mockGetExtensions.mockReturnValue([{ name: 'ext1' }, { name: 'ext2' }]);
-      const mockCwd = vi.spyOn(process, 'cwd').mockReturnValue('/test/dir');
-      await handleUninstall({ all: true });
-
-      expect(mockUninstallExtension).toHaveBeenCalledTimes(2);
-      expect(mockUninstallExtension).toHaveBeenCalledWith('ext1', false);
-      expect(mockUninstallExtension).toHaveBeenCalledWith('ext2', false);
-      expect(emitConsoleLog).toHaveBeenCalledWith(
-        'log',
-        'Extension "ext1" successfully uninstalled.',
-      );
-      expect(emitConsoleLog).toHaveBeenCalledWith(
-        'log',
-        'Extension "ext2" successfully uninstalled.',
-      );
-      mockCwd.mockRestore();
-    });
-
-    it('should log a message if no extensions are installed and --all flag is used', async () => {
-      mockLoadExtensions.mockResolvedValue(undefined);
-      mockGetExtensions.mockReturnValue([]);
-      const mockCwd = vi.spyOn(process, 'cwd').mockReturnValue('/test/dir');
-      await handleUninstall({ all: true });
-
-      expect(mockUninstallExtension).not.toHaveBeenCalled();
-      expect(emitConsoleLog).toHaveBeenCalledWith(
-        'log',
-        'No extensions currently installed.',
-      );
-      mockCwd.mockRestore();
-    });
-
     it('should report errors for failed uninstalls but continue with others', async () => {
       mockLoadExtensions.mockResolvedValue(undefined);
       const mockCwd = vi.spyOn(process, 'cwd').mockReturnValue('/test/dir');
@@ -274,14 +236,13 @@ describe('extensions uninstall command', () => {
     const command = uninstallCommand;
 
     it('should have correct command and describe', () => {
-      expect(command.command).toBe('uninstall [names..]');
+      expect(command.command).toBe('uninstall <names..>');
       expect(command.describe).toBe('Uninstalls one or more extensions.');
     });
 
     describe('builder', () => {
       interface MockYargs {
         positional: Mock;
-        option: Mock;
         check: Mock;
       }
 
@@ -289,12 +250,11 @@ describe('extensions uninstall command', () => {
       beforeEach(() => {
         yargsMock = {
           positional: vi.fn().mockReturnThis(),
-          option: vi.fn().mockReturnThis(),
           check: vi.fn().mockReturnThis(),
         };
       });
 
-      it('should configure arguments and options', () => {
+      it('should configure positional argument', () => {
         (command.builder as (yargs: Argv) => Argv)(
           yargsMock as unknown as Argv,
         );
@@ -304,31 +264,18 @@ describe('extensions uninstall command', () => {
           type: 'string',
           array: true,
         });
-        expect(yargsMock.option).toHaveBeenCalledWith('all', {
-          type: 'boolean',
-          describe: 'Uninstall all installed extensions.',
-          default: false,
-        });
         expect(yargsMock.check).toHaveBeenCalled();
       });
 
-      it('check function should throw for missing names and no --all flag', () => {
+      it('check function should throw for missing names', () => {
         (command.builder as (yargs: Argv) => Argv)(
           yargsMock as unknown as Argv,
         );
         const checkCallback = yargsMock.check.mock.calls[0][0];
-        expect(() => checkCallback({ names: [], all: false })).toThrow(
-          'Please include at least one extension name to uninstall as a positional argument, or use the --all flag.',
+        expect(() => checkCallback({ names: [] })).toThrow(
+          'Please include at least one extension name to uninstall as a positional argument.',
         );
       });
-
-      it('check function should pass if --all flag is used even without names', () => {
-        (command.builder as (yargs: Argv) => Argv)(
-          yargsMock as unknown as Argv,
-        );
-        const checkCallback = yargsMock.check.mock.calls[0][0];
-        expect(() => checkCallback({ names: [], all: true })).not.toThrow();
-      });
     });
 
     it('handler should call handleUninstall', async () => {
@@ -336,17 +283,10 @@ describe('extensions uninstall command', () => {
       mockUninstallExtension.mockResolvedValue(undefined);
       const mockCwd = vi.spyOn(process, 'cwd').mockReturnValue('/test/dir');
       interface TestArgv {
-        names?: string[];
-        all?: boolean;
-        _: string[];
-        $0: string;
+        names: string[];
+        [key: string]: unknown;
       }
-      const argv: TestArgv = {
-        names: ['my-extension'],
-        all: false,
-        _: [],
-        $0: '',
-      };
+      const argv: TestArgv = { names: ['my-extension'], _: [], $0: '' };
       await (command.handler as unknown as (args: TestArgv) => Promise<void>)(
         argv,
       );

packages/cli/src/commands/extensions/uninstall.ts

@@ -14,8 +14,7 @@ import { promptForSetting } from '../../config/extensions/extensionSettings.js';
 import { exitCli } from '../utils.js';
 
 interface UninstallArgs {
-  names?: string[]; // can be extension names or source URLs.
-  all?: boolean;
+  names: string[]; // can be extension names or source URLs.
 }
 
 export async function handleUninstall(args: UninstallArgs) {
@@ -29,24 +28,8 @@ export async function handleUninstall(args: UninstallArgs) {
     });
     await extensionManager.loadExtensions();
 
-    let namesToUninstall: string[] = [];
-    if (args.all) {
-      namesToUninstall = extensionManager
-        .getExtensions()
-        .map((ext) => ext.name);
-    } else if (args.names) {
-      namesToUninstall = [...new Set(args.names)];
-    }
-
-    if (namesToUninstall.length === 0) {
-      if (args.all) {
-        debugLogger.log('No extensions currently installed.');
-      }
-      return;
-    }
-
     const errors: Array<{ name: string; error: string }> = [];
-    for (const name of namesToUninstall) {
+    for (const name of [...new Set(args.names)]) {
       try {
         await extensionManager.uninstallExtension(name, false);
         debugLogger.log(`Extension "${name}" successfully uninstalled.`);
@@ -68,7 +51,7 @@ export async function handleUninstall(args: UninstallArgs) {
 }
 
 export const uninstallCommand: CommandModule = {
-  command: 'uninstall [names..]',
+  command: 'uninstall <names..>',
   describe: 'Uninstalls one or more extensions.',
   builder: (yargs) =>
     yargs
@@ -78,15 +61,10 @@ export const uninstallCommand: CommandModule = {
         type: 'string',
         array: true,
       })
-      .option('all', {
-        type: 'boolean',
-        describe: 'Uninstall all installed extensions.',
-        default: false,
-      })
       .check((argv) => {
-        if (!argv.all && (!argv.names || argv.names.length === 0)) {
+        if (!argv.names || argv.names.length === 0) {
           throw new Error(
-            'Please include at least one extension name to uninstall as a positional argument, or use the --all flag.',
+            'Please include at least one extension name to uninstall as a positional argument.',
           );
         }
         return true;
@@ -94,9 +72,7 @@ export const uninstallCommand: CommandModule = {
   handler: async (argv) => {
     await handleUninstall({
       // eslint-disable-next-line @typescript-eslint/no-unsafe-type-assertion
-      names: argv['names'] as string[] | undefined,
-      // eslint-disable-next-line @typescript-eslint/no-unsafe-type-assertion
-      all: argv['all'] as boolean,
+      names: argv['names'] as string[],
     });
     await exitCli();
   },

packages/cli/src/commands/hooks/migrate.ts

@@ -79,7 +79,6 @@ function migrateClaudeHook(claudeHook: unknown): unknown {
     migrated['command'] = hook['command'];
 
     // Replace CLAUDE_PROJECT_DIR with GEMINI_PROJECT_DIR in command
-    // eslint-disable-next-line no-restricted-syntax
     if (typeof migrated['command'] === 'string') {
       migrated['command'] = migrated['command'].replace(
         /\$CLAUDE_PROJECT_DIR/g,
@@ -94,7 +93,6 @@ function migrateClaudeHook(claudeHook: unknown): unknown {
   }
 
   // Map timeout field (Claude uses seconds, Gemini uses seconds)
-  // eslint-disable-next-line no-restricted-syntax
   if ('timeout' in hook && typeof hook['timeout'] === 'number') {
     migrated['timeout'] = hook['timeout'];
   }
@@ -142,7 +140,6 @@ function migrateClaudeHooks(claudeConfig: unknown): Record<string, unknown> {
       // Transform matcher
       if (
         'matcher' in definition &&
-        // eslint-disable-next-line no-restricted-syntax
         typeof definition['matcher'] === 'string'
       ) {
         migratedDef['matcher'] = transformMatcher(definition['matcher']);

packages/cli/src/config/config.test.ts

@@ -116,16 +116,14 @@ vi.mock('@google/gemini-cli-core', async () => {
       (
         cwd,
         dirs,
+        debug,
         fileService,
         extensionLoader: ExtensionLoader,
-        _folderTrust,
-        _importFormat,
-        _fileFilteringOptions,
         _maxDirs,
       ) => {
-        const extensionPaths =
-          extensionLoader?.getExtensions?.()?.flatMap((e) => e.contextFiles) ||
-          [];
+        const extensionPaths = extensionLoader
+          .getExtensions()
+          .flatMap((e) => e.contextFiles);
         return Promise.resolve({
           memoryContent: extensionPaths.join(',') || '',
           fileCount: extensionPaths?.length || 0,
@@ -849,6 +847,7 @@ describe('Hierarchical Memory Loading (config.ts) - Placeholder Suite', () => {
     expect(ServerConfig.loadServerHierarchicalMemory).toHaveBeenCalledWith(
       expect.any(String),
       [],
+      false,
       expect.any(Object),
       expect.any(ExtensionManager),
       true,
@@ -877,6 +876,7 @@ describe('Hierarchical Memory Loading (config.ts) - Placeholder Suite', () => {
     expect(ServerConfig.loadServerHierarchicalMemory).toHaveBeenCalledWith(
       expect.any(String),
       [includeDir],
+      false,
       expect.any(Object),
       expect.any(ExtensionManager),
       true,
@@ -904,6 +904,7 @@ describe('Hierarchical Memory Loading (config.ts) - Placeholder Suite', () => {
     expect(ServerConfig.loadServerHierarchicalMemory).toHaveBeenCalledWith(
       expect.any(String),
       [],
+      false,
       expect.any(Object),
       expect.any(ExtensionManager),
       true,
@@ -952,6 +953,12 @@ describe('mergeMcpServers', () => {
 });
 
 describe('mergeExcludeTools', () => {
+  const defaultExcludes = new Set([
+    SHELL_TOOL_NAME,
+    EDIT_TOOL_NAME,
+    WRITE_FILE_TOOL_NAME,
+    WEB_FETCH_TOOL_NAME,
+  ]);
   const originalIsTTY = process.stdin.isTTY;
 
   beforeEach(() => {
@@ -1073,7 +1080,9 @@ describe('mergeExcludeTools', () => {
     process.argv = ['node', 'script.js', '-p', 'test'];
     const argv = await parseArguments(createTestMergedSettings());
     const config = await loadCliConfig(settings, 'test-session', argv);
-    expect(config.getExcludeTools()).toEqual(new Set([ASK_USER_TOOL_NAME]));
+    expect(config.getExcludeTools()).toEqual(
+      new Set([...defaultExcludes, ASK_USER_TOOL_NAME]),
+    );
   });
 
   it('should handle settings with excludeTools but no extensions', async () => {
@@ -1154,9 +1163,9 @@ describe('Approval mode tool exclusion logic', () => {
     const config = await loadCliConfig(settings, 'test-session', argv);
 
     const excludedTools = config.getExcludeTools();
-    expect(excludedTools).not.toContain(SHELL_TOOL_NAME);
-    expect(excludedTools).not.toContain(EDIT_TOOL_NAME);
-    expect(excludedTools).not.toContain(WRITE_FILE_TOOL_NAME);
+    expect(excludedTools).toContain(SHELL_TOOL_NAME);
+    expect(excludedTools).toContain(EDIT_TOOL_NAME);
+    expect(excludedTools).toContain(WRITE_FILE_TOOL_NAME);
     expect(excludedTools).toContain(ASK_USER_TOOL_NAME);
   });
 
@@ -1175,9 +1184,9 @@ describe('Approval mode tool exclusion logic', () => {
     const config = await loadCliConfig(settings, 'test-session', argv);
 
     const excludedTools = config.getExcludeTools();
-    expect(excludedTools).not.toContain(SHELL_TOOL_NAME);
-    expect(excludedTools).not.toContain(EDIT_TOOL_NAME);
-    expect(excludedTools).not.toContain(WRITE_FILE_TOOL_NAME);
+    expect(excludedTools).toContain(SHELL_TOOL_NAME);
+    expect(excludedTools).toContain(EDIT_TOOL_NAME);
+    expect(excludedTools).toContain(WRITE_FILE_TOOL_NAME);
     expect(excludedTools).toContain(ASK_USER_TOOL_NAME);
   });
 
@@ -1196,7 +1205,7 @@ describe('Approval mode tool exclusion logic', () => {
     const config = await loadCliConfig(settings, 'test-session', argv);
 
     const excludedTools = config.getExcludeTools();
-    expect(excludedTools).not.toContain(SHELL_TOOL_NAME);
+    expect(excludedTools).toContain(SHELL_TOOL_NAME);
     expect(excludedTools).not.toContain(EDIT_TOOL_NAME);
     expect(excludedTools).not.toContain(WRITE_FILE_TOOL_NAME);
     expect(excludedTools).toContain(ASK_USER_TOOL_NAME);
@@ -1242,9 +1251,9 @@ describe('Approval mode tool exclusion logic', () => {
     const config = await loadCliConfig(settings, 'test-session', argv);
 
     const excludedTools = config.getExcludeTools();
-    expect(excludedTools).not.toContain(SHELL_TOOL_NAME);
-    expect(excludedTools).not.toContain(EDIT_TOOL_NAME);
-    expect(excludedTools).not.toContain(WRITE_FILE_TOOL_NAME);
+    expect(excludedTools).toContain(SHELL_TOOL_NAME);
+    expect(excludedTools).toContain(EDIT_TOOL_NAME);
+    expect(excludedTools).toContain(WRITE_FILE_TOOL_NAME);
     expect(excludedTools).toContain(ASK_USER_TOOL_NAME);
   });
 
@@ -1306,10 +1315,9 @@ describe('Approval mode tool exclusion logic', () => {
 
     const excludedTools = config.getExcludeTools();
     expect(excludedTools).toContain('custom_tool'); // From settings
-    expect(excludedTools).not.toContain(SHELL_TOOL_NAME); // No longer from approval mode
+    expect(excludedTools).toContain(SHELL_TOOL_NAME); // From approval mode
     expect(excludedTools).not.toContain(EDIT_TOOL_NAME); // Should be allowed in auto_edit
     expect(excludedTools).not.toContain(WRITE_FILE_TOOL_NAME); // Should be allowed in auto_edit
-    expect(excludedTools).toContain(ASK_USER_TOOL_NAME);
   });
 
   it('should throw an error if YOLO mode is attempted when disableYoloMode is true', async () => {
@@ -2156,9 +2164,9 @@ describe('loadCliConfig tool exclusions', () => {
       'test-session',
       argv,
     );
-    expect(config.getExcludeTools()).not.toContain('run_shell_command');
-    expect(config.getExcludeTools()).not.toContain('replace');
-    expect(config.getExcludeTools()).not.toContain('write_file');
+    expect(config.getExcludeTools()).toContain('run_shell_command');
+    expect(config.getExcludeTools()).toContain('replace');
+    expect(config.getExcludeTools()).toContain('write_file');
     expect(config.getExcludeTools()).toContain('ask_user');
   });
 
@@ -2196,7 +2204,7 @@ describe('loadCliConfig tool exclusions', () => {
     expect(config.getExcludeTools()).not.toContain(SHELL_TOOL_NAME);
   });
 
-  it('should not exclude web-fetch in non-interactive mode at config level', async () => {
+  it('should exclude web-fetch in non-interactive mode when not allowed', async () => {
     process.stdin.isTTY = false;
     process.argv = ['node', 'script.js', '-p', 'test'];
     const argv = await parseArguments(createTestMergedSettings());
@@ -2205,7 +2213,7 @@ describe('loadCliConfig tool exclusions', () => {
       'test-session',
       argv,
     );
-    expect(config.getExcludeTools()).not.toContain(WEB_FETCH_TOOL_NAME);
+    expect(config.getExcludeTools()).toContain(WEB_FETCH_TOOL_NAME);
   });
 
   it('should not exclude web-fetch in non-interactive mode when allowed', async () => {
@@ -3318,11 +3326,11 @@ describe('Policy Engine Integration in loadCliConfig', () => {
 
     await loadCliConfig(settings, 'test-session', argv);
 
-    // In non-interactive mode, only ask_user is excluded by default
+    // In non-interactive mode, ShellTool, etc. are excluded
     expect(ServerConfig.createPolicyEngineConfig).toHaveBeenCalledWith(
       expect.objectContaining({
         tools: expect.objectContaining({
-          exclude: expect.arrayContaining([ASK_USER_TOOL_NAME]),
+          exclude: expect.arrayContaining([SHELL_TOOL_NAME]),
         }),
       }),
       expect.anything(),

packages/cli/src/config/config.ts

@@ -19,11 +19,16 @@ import {
   DEFAULT_FILE_FILTERING_OPTIONS,
   DEFAULT_MEMORY_FILE_FILTERING_OPTIONS,
   FileDiscoveryService,
+  WRITE_FILE_TOOL_NAME,
+  SHELL_TOOL_NAMES,
+  SHELL_TOOL_NAME,
   resolveTelemetrySettings,
   FatalConfigError,
   getPty,
+  EDIT_TOOL_NAME,
   debugLogger,
   loadServerHierarchicalMemory,
+  WEB_FETCH_TOOL_NAME,
   ASK_USER_TOOL_NAME,
   getVersion,
   PREVIEW_GEMINI_MODEL_AUTO,
@@ -76,8 +81,7 @@ export interface CliArgs {
   policy: string[] | undefined;
   allowedMcpServerNames: string[] | undefined;
   allowedTools: string[] | undefined;
-  acp?: boolean;
-  experimentalAcp?: boolean;
+  experimentalAcp: boolean | undefined;
   extensions: string[] | undefined;
   listExtensions: boolean | undefined;
   resume: string | typeof RESUME_LATEST | undefined;
@@ -173,14 +177,9 @@ export async function parseArguments(
                 .filter(Boolean),
             ),
         })
-        .option('acp', {
-          type: 'boolean',
-          description: 'Starts the agent in ACP mode',
-        })
         .option('experimental-acp', {
           type: 'boolean',
-          description:
-            'Starts the agent in ACP mode (deprecated, use --acp instead)',
+          description: 'Starts the agent in ACP mode',
         })
         .option('allowed-mcp-server-names', {
           type: 'array',
@@ -396,6 +395,36 @@ export async function parseArguments(
   return result as unknown as CliArgs;
 }
 
+/**
+ * Creates a filter function to determine if a tool should be excluded.
+ *
+ * In non-interactive mode, we want to disable tools that require user
+ * interaction to prevent the CLI from hanging. This function creates a predicate
+ * that returns `true` if a tool should be excluded.
+ *
+ * A tool is excluded if it's not in the `allowedToolsSet`. The shell tool
+ * has a special case: it's not excluded if any of its subcommands
+ * are in the `allowedTools` list.
+ *
+ * @param allowedTools A list of explicitly allowed tool names.
+ * @param allowedToolsSet A set of explicitly allowed tool names for quick lookups.
+ * @returns A function that takes a tool name and returns `true` if it should be excluded.
+ */
+function createToolExclusionFilter(
+  allowedTools: string[],
+  allowedToolsSet: Set<string>,
+) {
+  return (tool: string): boolean => {
+    if (tool === SHELL_TOOL_NAME) {
+      // If any of the allowed tools is ShellTool (even with subcommands), don't exclude it.
+      return !allowedTools.some((allowed) =>
+        SHELL_TOOL_NAMES.some((shellName) => allowed.startsWith(shellName)),
+      );
+    }
+    return !allowedToolsSet.has(tool);
+  };
+}
+
 export function isDebugMode(argv: CliArgs): boolean {
   return (
     argv.debug ||
@@ -499,6 +528,7 @@ export async function loadCliConfig(
       settings.context?.loadMemoryFromIncludeDirectories || false
         ? includeDirectories
         : [],
+      debugMode,
       fileService,
       extensionManager,
       trustedFolder,
@@ -602,20 +632,54 @@ export async function loadCliConfig(
   // -i/--prompt-interactive forces interactive mode with an initial prompt
   const interactive =
     !!argv.promptInteractive ||
-    !!argv.acp ||
     !!argv.experimentalAcp ||
     (!isHeadlessMode({ prompt: argv.prompt, query: argv.query }) &&
       !argv.isCommand);
 
   const allowedTools = argv.allowedTools || settings.tools?.allowed || [];
+  const allowedToolsSet = new Set(allowedTools);
 
   // In non-interactive mode, exclude tools that require a prompt.
   const extraExcludes: string[] = [];
   if (!interactive) {
-    // The Policy Engine natively handles headless safety by translating ASK_USER
-    // decisions to DENY. However, we explicitly block ask_user here to guarantee
-    // it can never be allowed via a high-priority policy rule when no human is present.
+    // ask_user requires user interaction and must be excluded in all
+    // non-interactive modes, regardless of the approval mode.
     extraExcludes.push(ASK_USER_TOOL_NAME);
+
+    const defaultExcludes = [
+      SHELL_TOOL_NAME,
+      EDIT_TOOL_NAME,
+      WRITE_FILE_TOOL_NAME,
+      WEB_FETCH_TOOL_NAME,
+    ];
+    const autoEditExcludes = [SHELL_TOOL_NAME];
+
+    const toolExclusionFilter = createToolExclusionFilter(
+      allowedTools,
+      allowedToolsSet,
+    );
+
+    switch (approvalMode) {
+      case ApprovalMode.PLAN:
+        // In plan non-interactive mode, all tools that require approval are excluded.
+        // TODO(#16625): Replace this default exclusion logic with specific rules for plan mode.
+        extraExcludes.push(...defaultExcludes.filter(toolExclusionFilter));
+        break;
+      case ApprovalMode.DEFAULT:
+        // In default non-interactive mode, all tools that require approval are excluded.
+        extraExcludes.push(...defaultExcludes.filter(toolExclusionFilter));
+        break;
+      case ApprovalMode.AUTO_EDIT:
+        // In auto-edit non-interactive mode, only tools that still require a prompt are excluded.
+        extraExcludes.push(...autoEditExcludes.filter(toolExclusionFilter));
+        break;
+      case ApprovalMode.YOLO:
+        // No extra excludes for YOLO mode.
+        break;
+      default:
+        // This should never happen due to validation earlier, but satisfies the linter
+        break;
+    }
   }
 
   const excludeTools = mergeExcludeTools(settings, extraExcludes);
@@ -694,7 +758,6 @@ export async function loadCliConfig(
   }
 
   return new Config({
-    acpMode: !!argv.acp || !!argv.experimentalAcp,
     sessionId,
     clientVersion: await getVersion(),
     embeddingModel: DEFAULT_GEMINI_EMBEDDING_MODEL,
@@ -758,7 +821,7 @@ export async function loadCliConfig(
     bugCommand: settings.advanced?.bugCommand,
     model: resolvedModel,
     maxSessionTurns: settings.model?.maxSessionTurns,
-
+    experimentalZedIntegration: argv.experimentalAcp || false,
     listExtensions: argv.listExtensions || false,
     listSessions: argv.listSessions || false,
     deleteSession: argv.deleteSession,
@@ -805,7 +868,6 @@ export async function loadCliConfig(
     fakeResponses: argv.fakeResponses,
     recordResponses: argv.recordResponses,
     retryFetchErrors: settings.general?.retryFetchErrors,
-    billing: settings.billing,
     maxAttempts: settings.general?.maxAttempts,
     ptyInfo: ptyInfo?.name,
     disableLLMCorrection: settings.tools?.disableLLMCorrection,

packages/cli/src/config/extension-manager.test.ts

@@ -12,13 +12,6 @@ import { ExtensionManager } from './extension-manager.js';
 import { createTestMergedSettings } from './settings.js';
 import { createExtension } from '../test-utils/createExtension.js';
 import { EXTENSIONS_DIRECTORY_NAME } from './extensions/variables.js';
-import {
-  TrustLevel,
-  loadTrustedFolders,
-  isWorkspaceTrusted,
-} from './trustedFolders.js';
-import { getRealPath } from '@google/gemini-cli-core';
-import type { MergedSettings } from './settings.js';
 
 const mockHomedir = vi.hoisted(() => vi.fn(() => '/tmp/mock-home'));
 
@@ -192,157 +185,4 @@ describe('ExtensionManager', () => {
       fs.rmSync(externalDir, { recursive: true, force: true });
     });
   });
-
-  describe('symlink handling', () => {
-    let extensionDir: string;
-    let symlinkDir: string;
-
-    beforeEach(() => {
-      extensionDir = path.join(tempHomeDir, 'extension');
-      symlinkDir = path.join(tempHomeDir, 'symlink-ext');
-
-      fs.mkdirSync(extensionDir, { recursive: true });
-
-      fs.writeFileSync(
-        path.join(extensionDir, 'gemini-extension.json'),
-        JSON.stringify({ name: 'test-ext', version: '1.0.0' }),
-      );
-
-      fs.symlinkSync(extensionDir, symlinkDir, 'dir');
-    });
-
-    it('preserves symlinks in installMetadata.source when linking', async () => {
-      const manager = new ExtensionManager({
-        workspaceDir: tempWorkspaceDir,
-        settings: {
-          security: {
-            folderTrust: { enabled: false }, // Disable trust for simplicity in this test
-          },
-          experimental: { extensionConfig: false },
-          admin: { extensions: { enabled: true }, mcp: { enabled: true } },
-          hooksConfig: { enabled: true },
-        } as unknown as MergedSettings,
-        requestConsent: () => Promise.resolve(true),
-        requestSetting: null,
-      });
-
-      // Trust the workspace to allow installation
-      const trustedFolders = loadTrustedFolders();
-      await trustedFolders.setValue(tempWorkspaceDir, TrustLevel.TRUST_FOLDER);
-
-      const installMetadata = {
-        source: symlinkDir,
-        type: 'link' as const,
-      };
-
-      await manager.loadExtensions();
-      const extension = await manager.installOrUpdateExtension(installMetadata);
-
-      // Desired behavior: it preserves symlinks (if they were absolute or relative as provided)
-      expect(extension.installMetadata?.source).toBe(symlinkDir);
-    });
-
-    it('works with the new install command logic (preserves symlink but trusts real path)', async () => {
-      // This simulates the logic in packages/cli/src/commands/extensions/install.ts
-      const absolutePath = path.resolve(symlinkDir);
-      const realPath = getRealPath(absolutePath);
-
-      const settings = {
-        security: {
-          folderTrust: { enabled: true },
-        },
-        experimental: { extensionConfig: false },
-        admin: { extensions: { enabled: true }, mcp: { enabled: true } },
-        hooksConfig: { enabled: true },
-      } as unknown as MergedSettings;
-
-      // Trust the REAL path
-      const trustedFolders = loadTrustedFolders();
-      await trustedFolders.setValue(realPath, TrustLevel.TRUST_FOLDER);
-
-      // Check trust of the symlink path
-      const trustResult = isWorkspaceTrusted(settings, absolutePath);
-      expect(trustResult.isTrusted).toBe(true);
-
-      const manager = new ExtensionManager({
-        workspaceDir: tempWorkspaceDir,
-        settings,
-        requestConsent: () => Promise.resolve(true),
-        requestSetting: null,
-      });
-
-      const installMetadata = {
-        source: absolutePath,
-        type: 'link' as const,
-      };
-
-      await manager.loadExtensions();
-      const extension = await manager.installOrUpdateExtension(installMetadata);
-
-      expect(extension.installMetadata?.source).toBe(absolutePath);
-      expect(extension.installMetadata?.source).not.toBe(realPath);
-    });
-
-    it('enforces allowedExtensions using the real path', async () => {
-      const absolutePath = path.resolve(symlinkDir);
-      const realPath = getRealPath(absolutePath);
-
-      const settings = {
-        security: {
-          folderTrust: { enabled: false },
-          // Only allow the real path, not the symlink path
-          allowedExtensions: [realPath.replace(/\\/g, '\\\\')],
-        },
-        experimental: { extensionConfig: false },
-        admin: { extensions: { enabled: true }, mcp: { enabled: true } },
-        hooksConfig: { enabled: true },
-      } as unknown as MergedSettings;
-
-      const manager = new ExtensionManager({
-        workspaceDir: tempWorkspaceDir,
-        settings,
-        requestConsent: () => Promise.resolve(true),
-        requestSetting: null,
-      });
-
-      const installMetadata = {
-        source: absolutePath,
-        type: 'link' as const,
-      };
-
-      await manager.loadExtensions();
-      // This should pass because realPath is allowed
-      const extension = await manager.installOrUpdateExtension(installMetadata);
-      expect(extension.name).toBe('test-ext');
-
-      // Now try with a settings that only allows the symlink path string
-      const settingsOnlySymlink = {
-        security: {
-          folderTrust: { enabled: false },
-          // Only allow the symlink path string explicitly
-          allowedExtensions: [absolutePath.replace(/\\/g, '\\\\')],
-        },
-        experimental: { extensionConfig: false },
-        admin: { extensions: { enabled: true }, mcp: { enabled: true } },
-        hooksConfig: { enabled: true },
-      } as unknown as MergedSettings;
-
-      const manager2 = new ExtensionManager({
-        workspaceDir: tempWorkspaceDir,
-        settings: settingsOnlySymlink,
-        requestConsent: () => Promise.resolve(true),
-        requestSetting: null,
-      });
-
-      // This should FAIL because it checks the real path against the pattern
-      // (Unless symlinkDir === extensionDir, which shouldn't happen in this test setup)
-      if (absolutePath !== realPath) {
-        await expect(
-          manager2.installOrUpdateExtension(installMetadata),
-        ).rejects.toThrow(
-          /is not allowed by the "allowedExtensions" security setting/,
-        );
-      }
-    });
-  });
 });

packages/cli/src/config/extension-manager.ts

@@ -161,9 +161,7 @@ export class ExtensionManager extends ExtensionLoader {
       const extensionAllowed = this.settings.security?.allowedExtensions.some(
         (pattern) => {
           try {
-            return new RegExp(pattern).test(
-              getRealPath(installMetadata.source),
-            );
+            return new RegExp(pattern).test(installMetadata.source);
           } catch (e) {
             throw new Error(
               `Invalid regex pattern in allowedExtensions setting: "${pattern}. Error: ${getErrorMessage(e)}`,
@@ -212,9 +210,11 @@ export class ExtensionManager extends ExtensionLoader {
       await fs.promises.mkdir(extensionsDir, { recursive: true });
 
       if (installMetadata.type === 'local' || installMetadata.type === 'link') {
-        installMetadata.source = path.isAbsolute(installMetadata.source)
-          ? installMetadata.source
-          : path.resolve(this.workspaceDir, installMetadata.source);
+        installMetadata.source = getRealPath(
+          path.isAbsolute(installMetadata.source)
+            ? installMetadata.source
+            : path.resolve(this.workspaceDir, installMetadata.source),
+        );
       }
 
       let tempDir: string | undefined;
@@ -262,7 +262,7 @@ Would you like to attempt to install via "git clone" instead?`,
         installMetadata.type === 'local' ||
         installMetadata.type === 'link'
       ) {
-        localSourcePath = getRealPath(installMetadata.source);
+        localSourcePath = installMetadata.source;
       } else {
         throw new Error(`Unsupported install type: ${installMetadata.type}`);
       }
@@ -638,9 +638,7 @@ Would you like to attempt to install via "git clone" instead?`,
       const extensionAllowed = this.settings.security?.allowedExtensions.some(
         (pattern) => {
           try {
-            return new RegExp(pattern).test(
-              getRealPath(installMetadata?.source ?? ''),
-            );
+            return new RegExp(pattern).test(installMetadata?.source);
           } catch (e) {
             throw new Error(
               `Invalid regex pattern in allowedExtensions setting: "${pattern}. Error: ${getErrorMessage(e)}`,

packages/cli/src/config/footerItems.test.ts

@@ -1,91 +0,0 @@
-/**
- * @license
- * Copyright 2026 Google LLC
- * SPDX-License-Identifier: Apache-2.0
- */
-
-import { describe, it, expect } from 'vitest';
-import { deriveItemsFromLegacySettings } from './footerItems.js';
-import { createMockSettings } from '../test-utils/settings.js';
-
-describe('deriveItemsFromLegacySettings', () => {
-  it('returns defaults when no legacy settings are customized', () => {
-    const settings = createMockSettings({
-      ui: { footer: { hideContextPercentage: true } },
-    }).merged;
-    const items = deriveItemsFromLegacySettings(settings);
-    expect(items).toEqual([
-      'workspace',
-      'git-branch',
-      'sandbox',
-      'model-name',
-      'quota',
-    ]);
-  });
-
-  it('removes workspace when hideCWD is true', () => {
-    const settings = createMockSettings({
-      ui: { footer: { hideCWD: true, hideContextPercentage: true } },
-    }).merged;
-    const items = deriveItemsFromLegacySettings(settings);
-    expect(items).not.toContain('workspace');
-  });
-
-  it('removes sandbox when hideSandboxStatus is true', () => {
-    const settings = createMockSettings({
-      ui: { footer: { hideSandboxStatus: true, hideContextPercentage: true } },
-    }).merged;
-    const items = deriveItemsFromLegacySettings(settings);
-    expect(items).not.toContain('sandbox');
-  });
-
-  it('removes model-name, context-used, and quota when hideModelInfo is true', () => {
-    const settings = createMockSettings({
-      ui: { footer: { hideModelInfo: true, hideContextPercentage: true } },
-    }).merged;
-    const items = deriveItemsFromLegacySettings(settings);
-    expect(items).not.toContain('model-name');
-    expect(items).not.toContain('context-used');
-    expect(items).not.toContain('quota');
-  });
-
-  it('includes context-used when hideContextPercentage is false', () => {
-    const settings = createMockSettings({
-      ui: { footer: { hideContextPercentage: false } },
-    }).merged;
-    const items = deriveItemsFromLegacySettings(settings);
-    expect(items).toContain('context-used');
-    // Should be after model-name
-    const modelIdx = items.indexOf('model-name');
-    const contextIdx = items.indexOf('context-used');
-    expect(contextIdx).toBe(modelIdx + 1);
-  });
-
-  it('includes memory-usage when showMemoryUsage is true', () => {
-    const settings = createMockSettings({
-      ui: { showMemoryUsage: true, footer: { hideContextPercentage: true } },
-    }).merged;
-    const items = deriveItemsFromLegacySettings(settings);
-    expect(items).toContain('memory-usage');
-  });
-
-  it('handles combination of settings', () => {
-    const settings = createMockSettings({
-      ui: {
-        showMemoryUsage: true,
-        footer: {
-          hideCWD: true,
-          hideModelInfo: true,
-          hideContextPercentage: false,
-        },
-      },
-    }).merged;
-    const items = deriveItemsFromLegacySettings(settings);
-    expect(items).toEqual([
-      'git-branch',
-      'sandbox',
-      'context-used',
-      'memory-usage',
-    ]);
-  });
-});

packages/cli/src/config/footerItems.ts

@@ -1,132 +0,0 @@
-/**
- * @license
- * Copyright 2026 Google LLC
- * SPDX-License-Identifier: Apache-2.0
- */
-
-import type { MergedSettings } from './settings.js';
-
-export const ALL_ITEMS = [
-  {
-    id: 'workspace',
-    header: 'workspace (/directory)',
-    description: 'Current working directory',
-  },
-  {
-    id: 'git-branch',
-    header: 'branch',
-    description: 'Current git branch name (not shown when unavailable)',
-  },
-  {
-    id: 'sandbox',
-    header: 'sandbox',
-    description: 'Sandbox type and trust indicator',
-  },
-  {
-    id: 'model-name',
-    header: '/model',
-    description: 'Current model identifier',
-  },
-  {
-    id: 'context-used',
-    header: 'context',
-    description: 'Percentage of context window used',
-  },
-  {
-    id: 'quota',
-    header: '/stats',
-    description: 'Remaining usage on daily limit (not shown when unavailable)',
-  },
-  {
-    id: 'memory-usage',
-    header: 'memory',
-    description: 'Memory used by the application',
-  },
-  {
-    id: 'session-id',
-    header: 'session',
-    description: 'Unique identifier for the current session',
-  },
-  {
-    id: 'code-changes',
-    header: 'diff',
-    description: 'Lines added/removed in the session (not shown when zero)',
-  },
-  {
-    id: 'token-count',
-    header: 'tokens',
-    description: 'Total tokens used in the session (not shown when zero)',
-  },
-] as const;
-
-export type FooterItemId = (typeof ALL_ITEMS)[number]['id'];
-
-export const DEFAULT_ORDER = [
-  'workspace',
-  'git-branch',
-  'sandbox',
-  'model-name',
-  'context-used',
-  'quota',
-  'memory-usage',
-  'session-id',
-  'code-changes',
-  'token-count',
-];
-
-export function deriveItemsFromLegacySettings(
-  settings: MergedSettings,
-): string[] {
-  const defaults = [
-    'workspace',
-    'git-branch',
-    'sandbox',
-    'model-name',
-    'quota',
-  ];
-  const items = [...defaults];
-
-  const remove = (arr: string[], id: string) => {
-    const idx = arr.indexOf(id);
-    if (idx !== -1) arr.splice(idx, 1);
-  };
-
-  if (settings.ui.footer.hideCWD) remove(items, 'workspace');
-  if (settings.ui.footer.hideSandboxStatus) remove(items, 'sandbox');
-  if (settings.ui.footer.hideModelInfo) {
-    remove(items, 'model-name');
-    remove(items, 'context-used');
-    remove(items, 'quota');
-  }
-  if (
-    !settings.ui.footer.hideContextPercentage &&
-    !items.includes('context-used')
-  ) {
-    const modelIdx = items.indexOf('model-name');
-    if (modelIdx !== -1) items.splice(modelIdx + 1, 0, 'context-used');
-    else items.push('context-used');
-  }
-  if (settings.ui.showMemoryUsage) items.push('memory-usage');
-
-  return items;
-}
-
-const VALID_IDS: Set<string> = new Set(ALL_ITEMS.map((i) => i.id));
-
-/**
- * Resolves the ordered list and selected set of footer items from settings.
- * Used by FooterConfigDialog to initialize and reset state.
- */
-export function resolveFooterState(settings: MergedSettings): {
-  orderedIds: string[];
-  selectedIds: Set<string>;
-} {
-  const source = (
-    settings.ui?.footer?.items ?? deriveItemsFromLegacySettings(settings)
-  ).filter((id: string) => VALID_IDS.has(id));
-  const others = DEFAULT_ORDER.filter((id) => !source.includes(id));
-  return {
-    orderedIds: [...source, ...others],
-    selectedIds: new Set(source),
-  };
-}

packages/cli/src/config/keyBindings.test.ts

@@ -58,6 +58,46 @@ describe('keyBindings config', () => {
       const config: KeyBindingConfig = defaultKeyBindings;
       expect(config[Command.HOME]).toBeDefined();
     });
+
+    it('should have correct specific bindings', () => {
+      // Verify navigation ignores shift
+      const navUp = defaultKeyBindings[Command.NAVIGATION_UP];
+      expect(navUp).toContainEqual({ key: 'up', shift: false });
+
+      const navDown = defaultKeyBindings[Command.NAVIGATION_DOWN];
+      expect(navDown).toContainEqual({ key: 'down', shift: false });
+
+      // Verify dialog navigation
+      const dialogNavUp = defaultKeyBindings[Command.DIALOG_NAVIGATION_UP];
+      expect(dialogNavUp).toContainEqual({ key: 'up', shift: false });
+      expect(dialogNavUp).toContainEqual({ key: 'k', shift: false });
+
+      const dialogNavDown = defaultKeyBindings[Command.DIALOG_NAVIGATION_DOWN];
+      expect(dialogNavDown).toContainEqual({ key: 'down', shift: false });
+      expect(dialogNavDown).toContainEqual({ key: 'j', shift: false });
+
+      // Verify physical home/end keys for cursor movement
+      expect(defaultKeyBindings[Command.HOME]).toContainEqual({
+        key: 'home',
+        ctrl: false,
+        shift: false,
+      });
+      expect(defaultKeyBindings[Command.END]).toContainEqual({
+        key: 'end',
+        ctrl: false,
+        shift: false,
+      });
+
+      // Verify physical home/end keys for scrolling
+      expect(defaultKeyBindings[Command.SCROLL_HOME]).toContainEqual({
+        key: 'home',
+        ctrl: true,
+      });
+      expect(defaultKeyBindings[Command.SCROLL_END]).toContainEqual({
+        key: 'end',
+        ctrl: true,
+      });
+    });
   });
 
   describe('command metadata', () => {

packages/cli/src/config/keyBindings.ts

@@ -134,12 +134,27 @@ export const defaultKeyBindings: KeyBindingConfig = {
   [Command.EXIT]: [{ key: 'd', ctrl: true }],
 
   // Cursor Movement
-  [Command.HOME]: [{ key: 'a', ctrl: true }, { key: 'home' }],
-  [Command.END]: [{ key: 'e', ctrl: true }, { key: 'end' }],
-  [Command.MOVE_UP]: [{ key: 'up' }],
-  [Command.MOVE_DOWN]: [{ key: 'down' }],
-  [Command.MOVE_LEFT]: [{ key: 'left' }],
-  [Command.MOVE_RIGHT]: [{ key: 'right' }, { key: 'f', ctrl: true }],
+  [Command.HOME]: [
+    { key: 'a', ctrl: true },
+    { key: 'home', shift: false, ctrl: false },
+  ],
+  [Command.END]: [
+    { key: 'e', ctrl: true },
+    { key: 'end', shift: false, ctrl: false },
+  ],
+  [Command.MOVE_UP]: [
+    { key: 'up', shift: false, alt: false, ctrl: false, cmd: false },
+  ],
+  [Command.MOVE_DOWN]: [
+    { key: 'down', shift: false, alt: false, ctrl: false, cmd: false },
+  ],
+  [Command.MOVE_LEFT]: [
+    { key: 'left', shift: false, alt: false, ctrl: false, cmd: false },
+  ],
+  [Command.MOVE_RIGHT]: [
+    { key: 'right', shift: false, alt: false, ctrl: false, cmd: false },
+    { key: 'f', ctrl: true },
+  ],
   [Command.MOVE_WORD_LEFT]: [
     { key: 'left', ctrl: true },
     { key: 'left', alt: true },
@@ -168,8 +183,8 @@ export const defaultKeyBindings: KeyBindingConfig = {
   [Command.DELETE_CHAR_LEFT]: [{ key: 'backspace' }, { key: 'h', ctrl: true }],
   [Command.DELETE_CHAR_RIGHT]: [{ key: 'delete' }, { key: 'd', ctrl: true }],
   [Command.UNDO]: [
-    { key: 'z', cmd: true },
-    { key: 'z', alt: true },
+    { key: 'z', cmd: true, shift: false },
+    { key: 'z', alt: true, shift: false },
   ],
   [Command.REDO]: [
     { key: 'z', ctrl: true, shift: true },
@@ -192,33 +207,56 @@ export const defaultKeyBindings: KeyBindingConfig = {
   [Command.PAGE_DOWN]: [{ key: 'pagedown' }],
 
   // History & Search
-  [Command.HISTORY_UP]: [{ key: 'p', ctrl: true }],
-  [Command.HISTORY_DOWN]: [{ key: 'n', ctrl: true }],
+  [Command.HISTORY_UP]: [{ key: 'p', shift: false, ctrl: true }],
+  [Command.HISTORY_DOWN]: [{ key: 'n', shift: false, ctrl: true }],
   [Command.REVERSE_SEARCH]: [{ key: 'r', ctrl: true }],
-  [Command.REWIND]: [{ key: 'double escape' }], // for documentation only
-  [Command.SUBMIT_REVERSE_SEARCH]: [{ key: 'return' }],
-  [Command.ACCEPT_SUGGESTION_REVERSE_SEARCH]: [{ key: 'tab' }],
+  [Command.REWIND]: [{ key: 'double escape' }],
+  [Command.SUBMIT_REVERSE_SEARCH]: [{ key: 'return', ctrl: false }],
+  [Command.ACCEPT_SUGGESTION_REVERSE_SEARCH]: [{ key: 'tab', shift: false }],
 
   // Navigation
-  [Command.NAVIGATION_UP]: [{ key: 'up' }],
-  [Command.NAVIGATION_DOWN]: [{ key: 'down' }],
+  [Command.NAVIGATION_UP]: [{ key: 'up', shift: false }],
+  [Command.NAVIGATION_DOWN]: [{ key: 'down', shift: false }],
   // Navigation shortcuts appropriate for dialogs where we do not need to accept
   // text input.
-  [Command.DIALOG_NAVIGATION_UP]: [{ key: 'up' }, { key: 'k' }],
-  [Command.DIALOG_NAVIGATION_DOWN]: [{ key: 'down' }, { key: 'j' }],
-  [Command.DIALOG_NEXT]: [{ key: 'tab' }],
+  [Command.DIALOG_NAVIGATION_UP]: [
+    { key: 'up', shift: false },
+    { key: 'k', shift: false },
+  ],
+  [Command.DIALOG_NAVIGATION_DOWN]: [
+    { key: 'down', shift: false },
+    { key: 'j', shift: false },
+  ],
+  [Command.DIALOG_NEXT]: [{ key: 'tab', shift: false }],
   [Command.DIALOG_PREV]: [{ key: 'tab', shift: true }],
 
   // Suggestions & Completions
-  [Command.ACCEPT_SUGGESTION]: [{ key: 'tab' }, { key: 'return' }],
-  [Command.COMPLETION_UP]: [{ key: 'up' }, { key: 'p', ctrl: true }],
-  [Command.COMPLETION_DOWN]: [{ key: 'down' }, { key: 'n', ctrl: true }],
+  [Command.ACCEPT_SUGGESTION]: [
+    { key: 'tab', shift: false },
+    { key: 'return', ctrl: false },
+  ],
+  [Command.COMPLETION_UP]: [
+    { key: 'up', shift: false },
+    { key: 'p', shift: false, ctrl: true },
+  ],
+  [Command.COMPLETION_DOWN]: [
+    { key: 'down', shift: false },
+    { key: 'n', shift: false, ctrl: true },
+  ],
   [Command.EXPAND_SUGGESTION]: [{ key: 'right' }],
   [Command.COLLAPSE_SUGGESTION]: [{ key: 'left' }],
 
   // Text Input
   // Must also exclude shift to allow shift+enter for newline
-  [Command.SUBMIT]: [{ key: 'return' }],
+  [Command.SUBMIT]: [
+    {
+      key: 'return',
+      shift: false,
+      alt: false,
+      ctrl: false,
+      cmd: false,
+    },
+  ],
   [Command.NEWLINE]: [
     { key: 'return', ctrl: true },
     { key: 'return', cmd: true },
@@ -245,17 +283,19 @@ export const defaultKeyBindings: KeyBindingConfig = {
   [Command.TOGGLE_BACKGROUND_SHELL_LIST]: [{ key: 'l', ctrl: true }],
   [Command.KILL_BACKGROUND_SHELL]: [{ key: 'k', ctrl: true }],
   [Command.UNFOCUS_BACKGROUND_SHELL]: [{ key: 'tab', shift: true }],
-  [Command.UNFOCUS_BACKGROUND_SHELL_LIST]: [{ key: 'tab' }],
-  [Command.SHOW_BACKGROUND_SHELL_UNFOCUS_WARNING]: [{ key: 'tab' }],
-  [Command.SHOW_SHELL_INPUT_UNFOCUS_WARNING]: [{ key: 'tab' }],
+  [Command.UNFOCUS_BACKGROUND_SHELL_LIST]: [{ key: 'tab', shift: false }],
+  [Command.SHOW_BACKGROUND_SHELL_UNFOCUS_WARNING]: [
+    { key: 'tab', shift: false },
+  ],
+  [Command.SHOW_SHELL_INPUT_UNFOCUS_WARNING]: [{ key: 'tab', shift: false }],
   [Command.BACKGROUND_SHELL_SELECT]: [{ key: 'return' }],
   [Command.BACKGROUND_SHELL_ESCAPE]: [{ key: 'escape' }],
   [Command.SHOW_MORE_LINES]: [{ key: 'o', ctrl: true }],
   [Command.EXPAND_PASTE]: [{ key: 'o', ctrl: true }],
-  [Command.FOCUS_SHELL_INPUT]: [{ key: 'tab' }],
+  [Command.FOCUS_SHELL_INPUT]: [{ key: 'tab', shift: false }],
   [Command.UNFOCUS_SHELL_INPUT]: [{ key: 'tab', shift: true }],
   [Command.CLEAR_SCREEN]: [{ key: 'l', ctrl: true }],
-  [Command.RESTART_APP]: [{ key: 'r' }, { key: 'r', shift: true }],
+  [Command.RESTART_APP]: [{ key: 'r' }],
   [Command.SUSPEND_APP]: [{ key: 'z', ctrl: true }],
 };
 

packages/cli/src/config/policy-engine.integration.test.ts

@@ -4,7 +4,7 @@
  * SPDX-License-Identifier: Apache-2.0
  */
 
-import { describe, it, expect, vi, beforeEach, afterEach } from 'vitest';
+import { describe, it, expect, vi } from 'vitest';
 import {
   ApprovalMode,
   PolicyDecision,
@@ -29,10 +29,6 @@ vi.mock('@google/gemini-cli-core', async (importOriginal) => {
 });
 
 describe('Policy Engine Integration Tests', () => {
-  beforeEach(() => vi.stubEnv('GEMINI_SYSTEM_MD', ''));
-
-  afterEach(() => vi.unstubAllEnvs());
-
   describe('Policy configuration produces valid PolicyEngine config', () => {
     it('should create a working PolicyEngine from basic settings', async () => {
       const settings: Settings = {
@@ -93,13 +89,13 @@ describe('Policy Engine Integration Tests', () => {
       // Tools from allowed server should be allowed
       // Tools from allowed server should be allowed
       expect(
-        (await engine.check({ name: 'mcp_allowed-server_tool1' }, undefined))
+        (await engine.check({ name: 'allowed-server__tool1' }, undefined))
           .decision,
       ).toBe(PolicyDecision.ALLOW);
       expect(
         (
           await engine.check(
-            { name: 'mcp_allowed-server_another_tool' },
+            { name: 'allowed-server__another_tool' },
             undefined,
           )
         ).decision,
@@ -107,13 +103,13 @@ describe('Policy Engine Integration Tests', () => {
 
       // Tools from trusted server should be allowed
       expect(
-        (await engine.check({ name: 'mcp_trusted-server_tool1' }, undefined))
+        (await engine.check({ name: 'trusted-server__tool1' }, undefined))
           .decision,
       ).toBe(PolicyDecision.ALLOW);
       expect(
         (
           await engine.check(
-            { name: 'mcp_trusted-server_special_tool' },
+            { name: 'trusted-server__special_tool' },
             undefined,
           )
         ).decision,
@@ -121,17 +117,17 @@ describe('Policy Engine Integration Tests', () => {
 
       // Tools from blocked server should be denied
       expect(
-        (await engine.check({ name: 'mcp_blocked-server_tool1' }, undefined))
+        (await engine.check({ name: 'blocked-server__tool1' }, undefined))
           .decision,
       ).toBe(PolicyDecision.DENY);
       expect(
-        (await engine.check({ name: 'mcp_blocked-server_any_tool' }, undefined))
+        (await engine.check({ name: 'blocked-server__any_tool' }, undefined))
           .decision,
       ).toBe(PolicyDecision.DENY);
 
       // Tools from unknown servers should use default
       expect(
-        (await engine.check({ name: 'mcp_unknown-server_tool' }, undefined))
+        (await engine.check({ name: 'unknown-server__tool' }, undefined))
           .decision,
       ).toBe(PolicyDecision.ASK_USER);
     });
@@ -151,16 +147,12 @@ describe('Policy Engine Integration Tests', () => {
 
       // ANY tool with a server name should be allowed
       expect(
-        (await engine.check({ name: 'mcp_mcp-server_tool' }, 'mcp-server'))
+        (await engine.check({ name: 'mcp-server__tool' }, 'mcp-server'))
           .decision,
       ).toBe(PolicyDecision.ALLOW);
       expect(
-        (
-          await engine.check(
-            { name: 'mcp_another-server_tool' },
-            'another-server',
-          )
-        ).decision,
+        (await engine.check({ name: 'another-server__tool' }, 'another-server'))
+          .decision,
       ).toBe(PolicyDecision.ALLOW);
 
       // Built-in tools should NOT be allowed by the MCP wildcard
@@ -175,7 +167,7 @@ describe('Policy Engine Integration Tests', () => {
           allowed: ['my-server'],
         },
         tools: {
-          exclude: ['mcp_my-server_dangerous-tool'],
+          exclude: ['my-server__dangerous-tool'],
         },
       };
 
@@ -188,24 +180,20 @@ describe('Policy Engine Integration Tests', () => {
       // MCP server allowed (priority 4.1) provides general allow for server
       // MCP server allowed (priority 4.1) provides general allow for server
       expect(
-        (await engine.check({ name: 'mcp_my-server_safe-tool' }, undefined))
+        (await engine.check({ name: 'my-server__safe-tool' }, undefined))
           .decision,
       ).toBe(PolicyDecision.ALLOW);
       // But specific tool exclude (priority 4.4) wins over server allow
       expect(
-        (
-          await engine.check(
-            { name: 'mcp_my-server_dangerous-tool' },
-            undefined,
-          )
-        ).decision,
+        (await engine.check({ name: 'my-server__dangerous-tool' }, undefined))
+          .decision,
       ).toBe(PolicyDecision.DENY);
     });
 
     it('should handle complex mixed configurations', async () => {
       const settings: Settings = {
         tools: {
-          allowed: ['custom-tool', 'mcp_my-server_special-tool'],
+          allowed: ['custom-tool', 'my-server__special-tool'],
           exclude: ['glob', 'dangerous-tool'],
         },
         mcp: {
@@ -250,21 +238,21 @@ describe('Policy Engine Integration Tests', () => {
         (await engine.check({ name: 'custom-tool' }, undefined)).decision,
       ).toBe(PolicyDecision.ALLOW);
       expect(
-        (await engine.check({ name: 'mcp_my-server_special-tool' }, undefined))
+        (await engine.check({ name: 'my-server__special-tool' }, undefined))
           .decision,
       ).toBe(PolicyDecision.ALLOW);
 
       // MCP server tools
       expect(
-        (await engine.check({ name: 'mcp_allowed-server_tool' }, undefined))
+        (await engine.check({ name: 'allowed-server__tool' }, undefined))
           .decision,
       ).toBe(PolicyDecision.ALLOW);
       expect(
-        (await engine.check({ name: 'mcp_trusted-server_tool' }, undefined))
+        (await engine.check({ name: 'trusted-server__tool' }, undefined))
           .decision,
       ).toBe(PolicyDecision.ALLOW);
       expect(
-        (await engine.check({ name: 'mcp_blocked-server_tool' }, undefined))
+        (await engine.check({ name: 'blocked-server__tool' }, undefined))
           .decision,
       ).toBe(PolicyDecision.DENY);
 
@@ -491,7 +479,7 @@ describe('Policy Engine Integration Tests', () => {
       expect(blockedToolRule?.priority).toBe(4.4); // Command line exclude
 
       const blockedServerRule = rules.find(
-        (r) => r.toolName === 'mcp_blocked-server_*',
+        (r) => r.toolName === 'blocked-server__*',
       );
       expect(blockedServerRule?.priority).toBe(4.9); // MCP server exclude
 
@@ -501,13 +489,11 @@ describe('Policy Engine Integration Tests', () => {
       expect(specificToolRule?.priority).toBe(4.3); // Command line allow
 
       const trustedServerRule = rules.find(
-        (r) => r.toolName === 'mcp_trusted-server_*',
+        (r) => r.toolName === 'trusted-server__*',
       );
       expect(trustedServerRule?.priority).toBe(4.2); // MCP trusted server
 
-      const mcpServerRule = rules.find(
-        (r) => r.toolName === 'mcp_mcp-server_*',
-      );
+      const mcpServerRule = rules.find((r) => r.toolName === 'mcp-server__*');
       expect(mcpServerRule?.priority).toBe(4.1); // MCP allowed server
 
       const readOnlyToolRule = rules.find((r) => r.toolName === 'glob');
@@ -519,19 +505,18 @@ describe('Policy Engine Integration Tests', () => {
         (await engine.check({ name: 'blocked-tool' }, undefined)).decision,
       ).toBe(PolicyDecision.DENY);
       expect(
-        (await engine.check({ name: 'mcp_blocked-server_any' }, undefined))
+        (await engine.check({ name: 'blocked-server__any' }, undefined))
           .decision,
       ).toBe(PolicyDecision.DENY);
       expect(
         (await engine.check({ name: 'specific-tool' }, undefined)).decision,
       ).toBe(PolicyDecision.ALLOW);
       expect(
-        (await engine.check({ name: 'mcp_trusted-server_any' }, undefined))
+        (await engine.check({ name: 'trusted-server__any' }, undefined))
           .decision,
       ).toBe(PolicyDecision.ALLOW);
       expect(
-        (await engine.check({ name: 'mcp_mcp-server_any' }, undefined))
-          .decision,
+        (await engine.check({ name: 'mcp-server__any' }, undefined)).decision,
       ).toBe(PolicyDecision.ALLOW);
       expect((await engine.check({ name: 'glob' }, undefined)).decision).toBe(
         PolicyDecision.ALLOW,
@@ -560,7 +545,7 @@ describe('Policy Engine Integration Tests', () => {
 
       // Exclusion (195) should win over trust (90)
       expect(
-        (await engine.check({ name: 'mcp_conflicted-server_tool' }, undefined))
+        (await engine.check({ name: 'conflicted-server__tool' }, undefined))
           .decision,
       ).toBe(PolicyDecision.DENY);
     });
@@ -571,7 +556,7 @@ describe('Policy Engine Integration Tests', () => {
           excluded: ['my-server'], // Priority 195 - DENY
         },
         tools: {
-          allowed: ['mcp_my-server_special-tool'], // Priority 100 - ALLOW
+          allowed: ['my-server__special-tool'], // Priority 100 - ALLOW
         },
       };
 
@@ -584,11 +569,11 @@ describe('Policy Engine Integration Tests', () => {
       // Server exclusion (195) wins over specific tool allow (100)
       // This might be counterintuitive but follows the priority system
       expect(
-        (await engine.check({ name: 'mcp_my-server_special-tool' }, undefined))
+        (await engine.check({ name: 'my-server__special-tool' }, undefined))
           .decision,
       ).toBe(PolicyDecision.DENY);
       expect(
-        (await engine.check({ name: 'mcp_my-server_other-tool' }, undefined))
+        (await engine.check({ name: 'my-server__other-tool' }, undefined))
           .decision,
       ).toBe(PolicyDecision.DENY);
     });
@@ -658,13 +643,13 @@ describe('Policy Engine Integration Tests', () => {
       const tool3Rule = rules.find((r) => r.toolName === 'tool3');
       expect(tool3Rule?.priority).toBe(4.4); // Excluded tools (user tier)
 
-      const server2Rule = rules.find((r) => r.toolName === 'mcp_server2_*');
+      const server2Rule = rules.find((r) => r.toolName === 'server2__*');
       expect(server2Rule?.priority).toBe(4.9); // Excluded servers (user tier)
 
       const tool1Rule = rules.find((r) => r.toolName === 'tool1');
       expect(tool1Rule?.priority).toBe(4.3); // Allowed tools (user tier)
 
-      const server1Rule = rules.find((r) => r.toolName === 'mcp_server1_*');
+      const server1Rule = rules.find((r) => r.toolName === 'server1__*');
       expect(server1Rule?.priority).toBe(4.1); // Allowed servers (user tier)
 
       const globRule = rules.find((r) => r.toolName === 'glob');

packages/cli/src/config/policy.test.ts

@@ -183,6 +183,7 @@ describe('resolveWorkspacePolicyState', () => {
       setAutoAcceptWorkspacePolicies(originalValue);
     }
   });
+
   it('should not return workspace policies if cwd is the home directory', async () => {
     const policiesDir = path.join(tempDir, '.gemini', 'policies');
     fs.mkdirSync(policiesDir, { recursive: true });

packages/cli/src/config/sandboxConfig.test.ts

@@ -97,7 +97,7 @@ describe('loadSandboxConfig', () => {
     it('should throw if GEMINI_SANDBOX is an invalid command', async () => {
       process.env['GEMINI_SANDBOX'] = 'invalid-command';
       await expect(loadSandboxConfig({}, {})).rejects.toThrow(
-        "Invalid sandbox command 'invalid-command'. Must be one of docker, podman, sandbox-exec, runsc, lxc",
+        "Invalid sandbox command 'invalid-command'. Must be one of docker, podman, sandbox-exec",
       );
     });
 
@@ -108,22 +108,6 @@ describe('loadSandboxConfig', () => {
         "Missing sandbox command 'docker' (from GEMINI_SANDBOX)",
       );
     });
-
-    it('should use lxc if GEMINI_SANDBOX=lxc and it exists', async () => {
-      process.env['GEMINI_SANDBOX'] = 'lxc';
-      mockedCommandExistsSync.mockReturnValue(true);
-      const config = await loadSandboxConfig({}, {});
-      expect(config).toEqual({ command: 'lxc', image: 'default/image' });
-      expect(mockedCommandExistsSync).toHaveBeenCalledWith('lxc');
-    });
-
-    it('should throw if GEMINI_SANDBOX=lxc but lxc command does not exist', async () => {
-      process.env['GEMINI_SANDBOX'] = 'lxc';
-      mockedCommandExistsSync.mockReturnValue(false);
-      await expect(loadSandboxConfig({}, {})).rejects.toThrow(
-        "Missing sandbox command 'lxc' (from GEMINI_SANDBOX)",
-      );
-    });
   });
 
   describe('with sandbox: true', () => {
@@ -194,7 +178,7 @@ describe('loadSandboxConfig', () => {
       await expect(
         loadSandboxConfig({}, { sandbox: 'invalid-command' }),
       ).rejects.toThrow(
-        "Invalid sandbox command 'invalid-command'. Must be one of docker, podman, sandbox-exec, runsc, lxc",
+        "Invalid sandbox command 'invalid-command'. Must be one of docker, podman, sandbox-exec",
       );
     });
   });
@@ -247,92 +231,4 @@ describe('loadSandboxConfig', () => {
       },
     );
   });
-
-  describe('with sandbox: runsc (gVisor)', () => {
-    beforeEach(() => {
-      mockedOsPlatform.mockReturnValue('linux');
-      mockedCommandExistsSync.mockReturnValue(true);
-    });
-
-    it('should use runsc via CLI argument on Linux', async () => {
-      const config = await loadSandboxConfig({}, { sandbox: 'runsc' });
-
-      expect(config).toEqual({ command: 'runsc', image: 'default/image' });
-      expect(mockedCommandExistsSync).toHaveBeenCalledWith('runsc');
-      expect(mockedCommandExistsSync).toHaveBeenCalledWith('docker');
-    });
-
-    it('should use runsc via GEMINI_SANDBOX environment variable', async () => {
-      process.env['GEMINI_SANDBOX'] = 'runsc';
-      const config = await loadSandboxConfig({}, {});
-
-      expect(config).toEqual({ command: 'runsc', image: 'default/image' });
-      expect(mockedCommandExistsSync).toHaveBeenCalledWith('runsc');
-      expect(mockedCommandExistsSync).toHaveBeenCalledWith('docker');
-    });
-
-    it('should use runsc via settings file', async () => {
-      const config = await loadSandboxConfig(
-        { tools: { sandbox: 'runsc' } },
-        {},
-      );
-
-      expect(config).toEqual({ command: 'runsc', image: 'default/image' });
-      expect(mockedCommandExistsSync).toHaveBeenCalledWith('runsc');
-      expect(mockedCommandExistsSync).toHaveBeenCalledWith('docker');
-    });
-
-    it('should prioritize GEMINI_SANDBOX over CLI and settings', async () => {
-      process.env['GEMINI_SANDBOX'] = 'runsc';
-      const config = await loadSandboxConfig(
-        { tools: { sandbox: 'docker' } },
-        { sandbox: 'podman' },
-      );
-
-      expect(config).toEqual({ command: 'runsc', image: 'default/image' });
-    });
-
-    it('should reject runsc on macOS (Linux-only)', async () => {
-      mockedOsPlatform.mockReturnValue('darwin');
-
-      await expect(loadSandboxConfig({}, { sandbox: 'runsc' })).rejects.toThrow(
-        'gVisor (runsc) sandboxing is only supported on Linux',
-      );
-    });
-
-    it('should reject runsc on Windows (Linux-only)', async () => {
-      mockedOsPlatform.mockReturnValue('win32');
-
-      await expect(loadSandboxConfig({}, { sandbox: 'runsc' })).rejects.toThrow(
-        'gVisor (runsc) sandboxing is only supported on Linux',
-      );
-    });
-
-    it('should throw if runsc binary not found', async () => {
-      mockedCommandExistsSync.mockReturnValue(false);
-
-      await expect(loadSandboxConfig({}, { sandbox: 'runsc' })).rejects.toThrow(
-        "Missing sandbox command 'runsc' (from GEMINI_SANDBOX)",
-      );
-    });
-
-    it('should throw if Docker not available (runsc requires Docker)', async () => {
-      mockedCommandExistsSync.mockImplementation((cmd) => cmd === 'runsc');
-
-      await expect(loadSandboxConfig({}, { sandbox: 'runsc' })).rejects.toThrow(
-        "runsc (gVisor) requires Docker. Install Docker, or use sandbox: 'docker'.",
-      );
-    });
-
-    it('should NOT auto-detect runsc when both runsc and docker available', async () => {
-      mockedCommandExistsSync.mockImplementation(
-        (cmd) => cmd === 'runsc' || cmd === 'docker',
-      );
-
-      const config = await loadSandboxConfig({}, { sandbox: true });
-
-      expect(config?.command).toBe('docker');
-      expect(config?.command).not.toBe('runsc');
-    });
-  });
 });

packages/cli/src/config/sandboxConfig.ts

@@ -27,8 +27,6 @@ const VALID_SANDBOX_COMMANDS: ReadonlyArray<SandboxConfig['command']> = [
   'docker',
   'podman',
   'sandbox-exec',
-  'runsc',
-  'lxc',
 ];
 
 function isSandboxCommand(value: string): value is SandboxConfig['command'] {
@@ -65,30 +63,17 @@ function getSandboxCommand(
         )}`,
       );
     }
-    // runsc (gVisor) is only supported on Linux
-    if (sandbox === 'runsc' && os.platform() !== 'linux') {
-      throw new FatalSandboxError(
-        'gVisor (runsc) sandboxing is only supported on Linux',
-      );
-    }
     // confirm that specified command exists
-    if (!commandExists.sync(sandbox)) {
-      throw new FatalSandboxError(
-        `Missing sandbox command '${sandbox}' (from GEMINI_SANDBOX)`,
-      );
+    if (commandExists.sync(sandbox)) {
+      return sandbox;
     }
-    // runsc uses Docker with --runtime=runsc; both must be available (prioritize runsc when explicitly chosen)
-    if (sandbox === 'runsc' && !commandExists.sync('docker')) {
-      throw new FatalSandboxError(
-        "runsc (gVisor) requires Docker. Install Docker, or use sandbox: 'docker'.",
-      );
-    }
-    return sandbox;
+    throw new FatalSandboxError(
+      `Missing sandbox command '${sandbox}' (from GEMINI_SANDBOX)`,
+    );
   }
 
   // look for seatbelt, docker, or podman, in that order
   // for container-based sandboxing, require sandbox to be enabled explicitly
-  // note: runsc is NOT auto-detected, it must be explicitly specified
   if (os.platform() === 'darwin' && commandExists.sync('sandbox-exec')) {
     return 'sandbox-exec';
   } else if (commandExists.sync('docker') && sandbox === true) {
@@ -106,9 +91,6 @@ function getSandboxCommand(
   }
 
   return '';
-  // Note: 'lxc' is intentionally not auto-detected because it requires a
-  // pre-existing, running container managed by the user. Use
-  // GEMINI_SANDBOX=lxc or sandbox: "lxc" in settings to enable it.
 }
 
 export async function loadSandboxConfig(

packages/cli/src/config/settings.test.ts

@@ -2162,7 +2162,7 @@ describe('Settings Loading and Merging', () => {
       }
     });
 
-    it('should remove deprecated settings by default and prioritize new ones', () => {
+    it('should prioritize new settings over deprecated ones and respect removeDeprecated flag', () => {
       const userSettingsContent = {
         general: {
           disableAutoUpdate: true,
@@ -2177,41 +2177,13 @@ describe('Settings Loading and Merging', () => {
       };
 
       const loadedSettings = createMockSettings(userSettingsContent);
+
       const setValueSpy = vi.spyOn(loadedSettings, 'setValue');
 
-      // Default is now removeDeprecated = true
+      // 1. removeDeprecated = false (default)
       migrateDeprecatedSettings(loadedSettings);
 
-      // Should remove disableAutoUpdate and trust enableAutoUpdate: true
-      expect(setValueSpy).toHaveBeenCalledWith(SettingScope.User, 'general', {
-        enableAutoUpdate: true,
-      });
-
-      // Should remove disableFuzzySearch and trust enableFuzzySearch: false
-      expect(setValueSpy).toHaveBeenCalledWith(SettingScope.User, 'context', {
-        fileFiltering: { enableFuzzySearch: false },
-      });
-    });
-
-    it('should preserve deprecated settings when removeDeprecated is explicitly false', () => {
-      const userSettingsContent = {
-        general: {
-          disableAutoUpdate: true,
-          enableAutoUpdate: true,
-        },
-        context: {
-          fileFiltering: {
-            disableFuzzySearch: false,
-            enableFuzzySearch: false,
-          },
-        },
-      };
-
-      const loadedSettings = createMockSettings(userSettingsContent);
-
-      migrateDeprecatedSettings(loadedSettings, false);
-
-      // Should still have old settings since removeDeprecated = false
+      // Should still have old settings
       expect(
         loadedSettings.forScope(SettingScope.User).settings.general,
       ).toHaveProperty('disableAutoUpdate');
@@ -2222,6 +2194,19 @@ describe('Settings Loading and Merging', () => {
           }
         ).fileFiltering,
       ).toHaveProperty('disableFuzzySearch');
+
+      // 2. removeDeprecated = true
+      migrateDeprecatedSettings(loadedSettings, true);
+
+      // Should remove disableAutoUpdate and trust enableAutoUpdate: true
+      expect(setValueSpy).toHaveBeenCalledWith(SettingScope.User, 'general', {
+        enableAutoUpdate: true,
+      });
+
+      // Should remove disableFuzzySearch and trust enableFuzzySearch: false
+      expect(setValueSpy).toHaveBeenCalledWith(SettingScope.User, 'context', {
+        fileFiltering: { enableFuzzySearch: false },
+      });
     });
 
     it('should trigger migration automatically during loadSettings', () => {

packages/cli/src/config/settings.ts

@@ -20,8 +20,8 @@ import {
   type AdminControlsSettings,
 } from '@google/gemini-cli-core';
 import stripJsonComments from 'strip-json-comments';
-import { DefaultLight } from '../ui/themes/builtin/light/default-light.js';
-import { DefaultDark } from '../ui/themes/builtin/dark/default-dark.js';
+import { DefaultLight } from '../ui/themes/default-light.js';
+import { DefaultDark } from '../ui/themes/default.js';
 import { isWorkspaceTrusted } from './trustedFolders.js';
 import {
   type Settings,
@@ -796,13 +796,14 @@ export function loadSettings(
 /**
  * Migrates deprecated settings to their new counterparts.
  *
- * Deprecated settings are removed from settings files by default.
+ * TODO: After a couple of weeks (around early Feb 2026), we should start removing
+ * the deprecated settings from the settings files by default.
  *
  * @returns true if any changes were made and need to be saved.
  */
 export function migrateDeprecatedSettings(
   loadedSettings: LoadedSettings,
-  removeDeprecated = true,
+  removeDeprecated = false,
 ): boolean {
   let anyModified = false;
   const systemWarnings: Map<LoadableSettingScope, string[]> = new Map();

packages/cli/src/config/settingsSchema.test.ts

@@ -461,7 +461,7 @@ describe('SettingsSchema', () => {
       expect(gemmaModelRouter.category).toBe('Experimental');
       expect(gemmaModelRouter.default).toEqual({});
       expect(gemmaModelRouter.requiresRestart).toBe(true);
-      expect(gemmaModelRouter.showInDialog).toBe(false);
+      expect(gemmaModelRouter.showInDialog).toBe(true);
       expect(gemmaModelRouter.description).toBe(
         'Enable Gemma model router (experimental).',
       );
@@ -472,9 +472,9 @@ describe('SettingsSchema', () => {
       expect(enabled.category).toBe('Experimental');
       expect(enabled.default).toBe(false);
       expect(enabled.requiresRestart).toBe(true);
-      expect(enabled.showInDialog).toBe(false);
+      expect(enabled.showInDialog).toBe(true);
       expect(enabled.description).toBe(
-        'Enable the Gemma Model Router (experimental). Requires a local endpoint serving Gemma via the Gemini API using LiteRT-LM shim.',
+        'Enable the Gemma Model Router. Requires a local endpoint serving Gemma via the Gemini API using LiteRT-LM shim.',
       );
 
       const classifier = gemmaModelRouter.properties.classifier;

packages/cli/src/config/settingsSchema.ts

@@ -565,34 +565,14 @@ const SETTINGS_SCHEMA = {
         description: 'Settings for the footer.',
         showInDialog: false,
         properties: {
-          items: {
-            type: 'array',
-            label: 'Footer Items',
-            category: 'UI',
-            requiresRestart: false,
-            default: undefined as string[] | undefined,
-            description:
-              'List of item IDs to display in the footer. Rendered in order',
-            showInDialog: false,
-            items: { type: 'string' },
-          },
-          showLabels: {
-            type: 'boolean',
-            label: 'Show Footer Labels',
-            category: 'UI',
-            requiresRestart: false,
-            default: true,
-            description:
-              'Display a second line above the footer items with descriptive headers (e.g., /model).',
-            showInDialog: false,
-          },
           hideCWD: {
             type: 'boolean',
             label: 'Hide CWD',
             category: 'UI',
             requiresRestart: false,
             default: false,
-            description: 'Hide the current working directory in the footer.',
+            description:
+              'Hide the current working directory path in the footer.',
             showInDialog: true,
           },
           hideSandboxStatus: {
@@ -1169,7 +1149,7 @@ const SETTINGS_SCHEMA = {
         requiresRestart: false,
         default: false,
         description: oneLine`
-          Controls how /memory reload loads GEMINI.md files.
+          Controls how /memory refresh loads GEMINI.md files.
           When true, include directories are scanned; when false, only the current directory is used.
         `,
         showInDialog: true,
@@ -1256,8 +1236,7 @@ const SETTINGS_SCHEMA = {
         ref: 'BooleanOrString',
         description: oneLine`
           Sandbox execution environment.
-          Set to a boolean to enable or disable the sandbox, provide a string path to a sandbox profile,
-          or specify an explicit sandbox command (e.g., "docker", "podman", "lxc").
+          Set to a boolean to enable or disable the sandbox, or provide a string path to a sandbox profile.
         `,
         showInDialog: false,
       },
@@ -1832,7 +1811,7 @@ const SETTINGS_SCHEMA = {
         label: 'Task Tracker',
         category: 'Experimental',
         requiresRestart: true,
-        default: false,
+        default: true,
         description: 'Enable task tracker tools.',
         showInDialog: false,
       },
@@ -1863,7 +1842,7 @@ const SETTINGS_SCHEMA = {
         requiresRestart: true,
         default: {},
         description: 'Enable Gemma model router (experimental).',
-        showInDialog: false,
+        showInDialog: true,
         properties: {
           enabled: {
             type: 'boolean',
@@ -1872,8 +1851,8 @@ const SETTINGS_SCHEMA = {
             requiresRestart: true,
             default: false,
             description:
-              'Enable the Gemma Model Router (experimental). Requires a local endpoint serving Gemma via the Gemini API using LiteRT-LM shim.',
-            showInDialog: false,
+              'Enable the Gemma Model Router. Requires a local endpoint serving Gemma via the Gemini API using LiteRT-LM shim.',
+            showInDialog: true,
           },
           classifier: {
             type: 'object',

packages/cli/src/config/trustedFolders.test.ts

@@ -506,7 +506,7 @@ describe('Trusted Folders', () => {
         const realDir = path.join(tempDir, 'real');
         const symlinkDir = path.join(tempDir, 'symlink');
         fs.mkdirSync(realDir);
-        fs.symlinkSync(realDir, symlinkDir, 'dir');
+        fs.symlinkSync(realDir, symlinkDir);
 
         // Rule uses realpath
         const config = { [realDir]: TrustLevel.TRUST_FOLDER };

packages/cli/src/core/auth.test.ts

@@ -9,7 +9,6 @@ import { performInitialAuth } from './auth.js';
 import {
   type Config,
   ValidationRequiredError,
-  ProjectIdRequiredError,
   AuthType,
 } from '@google/gemini-cli-core';
 
@@ -117,22 +116,4 @@ describe('auth', () => {
       AuthType.LOGIN_WITH_GOOGLE,
     );
   });
-
-  it('should return ProjectIdRequiredError message without "Failed to login" prefix', async () => {
-    const projectIdError = new ProjectIdRequiredError();
-    vi.mocked(mockConfig.refreshAuth).mockRejectedValue(projectIdError);
-    const result = await performInitialAuth(
-      mockConfig,
-      AuthType.LOGIN_WITH_GOOGLE,
-    );
-    expect(result).toEqual({
-      authError:
-        'This account requires setting the GOOGLE_CLOUD_PROJECT or GOOGLE_CLOUD_PROJECT_ID env var. See https://goo.gle/gemini-cli-auth-docs#workspace-gca',
-      accountSuspensionInfo: null,
-    });
-    expect(result.authError).not.toContain('Failed to login');
-    expect(mockConfig.refreshAuth).toHaveBeenCalledWith(
-      AuthType.LOGIN_WITH_GOOGLE,
-    );
-  });
 });

packages/cli/src/core/auth.ts

@@ -10,7 +10,6 @@ import {
   getErrorMessage,
   ValidationRequiredError,
   isAccountSuspendedError,
-  ProjectIdRequiredError,
 } from '@google/gemini-cli-core';
 
 import type { AccountSuspensionInfo } from '../ui/contexts/UIStateContext.js';
@@ -55,14 +54,6 @@ export async function performInitialAuth(
         },
       };
     }
-    if (e instanceof ProjectIdRequiredError) {
-      // OAuth succeeded but account setup requires project ID
-      // Show the error message directly without "Failed to login" prefix
-      return {
-        authError: getErrorMessage(e),
-        accountSuspensionInfo: null,
-      };
-    }
     return {
       authError: `Failed to login. Message: ${getErrorMessage(e)}`,
       accountSuspensionInfo: null,

packages/cli/src/gemini.test.tsx

@@ -747,60 +747,6 @@ describe('gemini.tsx main function kitty protocol', () => {
     emitFeedbackSpy.mockRestore();
   });
 
-  it('should start normally with a warning when no sessions found for resume', async () => {
-    const { SessionSelector, SessionError } = await import(
-      './utils/sessionUtils.js'
-    );
-    vi.mocked(SessionSelector).mockImplementation(
-      () =>
-        ({
-          resolveSession: vi
-            .fn()
-            .mockRejectedValue(SessionError.noSessionsFound()),
-        }) as unknown as InstanceType<typeof SessionSelector>,
-    );
-
-    const processExitSpy = vi
-      .spyOn(process, 'exit')
-      .mockImplementation((code) => {
-        throw new MockProcessExitError(code);
-      });
-    const emitFeedbackSpy = vi.spyOn(coreEvents, 'emitFeedback');
-
-    vi.mocked(loadSettings).mockReturnValue(
-      createMockSettings({
-        merged: { advanced: {}, security: { auth: {} }, ui: { theme: 'test' } },
-        workspace: { settings: {} },
-        setValue: vi.fn(),
-        forScope: () => ({ settings: {}, originalSettings: {}, path: '' }),
-      }),
-    );
-
-    vi.mocked(parseArguments).mockResolvedValue({
-      promptInteractive: false,
-      resume: 'latest',
-    } as unknown as CliArgs);
-    vi.mocked(loadCliConfig).mockResolvedValue(
-      createMockConfig({
-        isInteractive: () => true,
-        getQuestion: () => '',
-        getSandbox: () => undefined,
-      }),
-    );
-
-    await main();
-
-    // Should NOT have crashed
-    expect(processExitSpy).not.toHaveBeenCalled();
-    // Should NOT have emitted a feedback error
-    expect(emitFeedbackSpy).not.toHaveBeenCalledWith(
-      'error',
-      expect.stringContaining('Error resuming session'),
-    );
-    processExitSpy.mockRestore();
-    emitFeedbackSpy.mockRestore();
-  });
-
   it.skip('should log error when cleanupExpiredSessions fails', async () => {
     const { cleanupExpiredSessions } = await import(
       './utils/sessionCleanup.js'
@@ -1013,18 +959,13 @@ describe('gemini.tsx main function exit codes', () => {
       resume: 'invalid-session',
     } as unknown as CliArgs);
 
-    vi.mock('./utils/sessionUtils.js', async (importOriginal) => {
-      const original =
-        await importOriginal<typeof import('./utils/sessionUtils.js')>();
-      return {
-        ...original,
-        SessionSelector: vi.fn().mockImplementation(() => ({
-          resolveSession: vi
-            .fn()
-            .mockRejectedValue(new Error('Session not found')),
-        })),
-      };
-    });
+    vi.mock('./utils/sessionUtils.js', () => ({
+      SessionSelector: vi.fn().mockImplementation(() => ({
+        resolveSession: vi
+          .fn()
+          .mockRejectedValue(new Error('Session not found')),
+      })),
+    }));
 
     process.env['GEMINI_API_KEY'] = 'test-key';
     try {

packages/cli/src/gemini.tsx

@@ -79,12 +79,12 @@ import {
   type InitializationResult,
 } from './core/initializer.js';
 import { validateAuthMethod } from './config/auth.js';
-import { runAcpClient } from './acp/acpClient.js';
+import { runZedIntegration } from './zed-integration/zedIntegration.js';
 import { validateNonInteractiveAuth } from './validateNonInterActiveAuth.js';
 import { checkForUpdates } from './ui/utils/updateCheck.js';
 import { handleAutoUpdate } from './utils/handleAutoUpdate.js';
 import { appEvents, AppEvent } from './utils/events.js';
-import { SessionError, SessionSelector } from './utils/sessionUtils.js';
+import { SessionSelector } from './utils/sessionUtils.js';
 import { SettingsContext } from './ui/contexts/SettingsContext.js';
 import { MouseProvider } from './ui/contexts/MouseContext.js';
 import { StreamingState } from './ui/types.js';
@@ -672,8 +672,8 @@ export async function main() {
       await getOauthClient(settings.merged.security.auth.selectedType, config);
     }
 
-    if (config.getAcpMode()) {
-      return runAcpClient(config, settings, argv);
+    if (config.getExperimentalZedIntegration()) {
+      return runZedIntegration(config, settings, argv);
     }
 
     let input = config.getQuestion();
@@ -706,24 +706,12 @@ export async function main() {
         // Use the existing session ID to continue recording to the same session
         config.setSessionId(resumedSessionData.conversation.sessionId);
       } catch (error) {
-        if (
-          error instanceof SessionError &&
-          error.code === 'NO_SESSIONS_FOUND'
-        ) {
-          // No sessions to resume — start a fresh session with a warning
-          startupWarnings.push({
-            id: 'resume-no-sessions',
-            message: error.message,
-            priority: WarningPriority.High,
-          });
-        } else {
-          coreEvents.emitFeedback(
-            'error',
-            `Error resuming session: ${error instanceof Error ? error.message : 'Unknown error'}`,
-          );
-          await runExitCleanup();
-          process.exit(ExitCodes.FATAL_INPUT_ERROR);
-        }
+        coreEvents.emitFeedback(
+          'error',
+          `Error resuming session: ${error instanceof Error ? error.message : 'Unknown error'}`,
+        );
+        await runExitCleanup();
+        process.exit(ExitCodes.FATAL_INPUT_ERROR);
       }
     }
 

packages/cli/src/gemini_cleanup.test.tsx

@@ -179,7 +179,7 @@ describe('gemini.tsx main function cleanup', () => {
     vi.restoreAllMocks();
   });
 
-  it.skip('should log error when cleanupExpiredSessions fails', async () => {
+  it('should log error when cleanupExpiredSessions fails', async () => {
     const { loadCliConfig, parseArguments } = await import(
       './config/config.js'
     );
@@ -216,7 +216,7 @@ describe('gemini.tsx main function cleanup', () => {
       getMcpServers: () => ({}),
       getMcpClientManager: vi.fn(),
       getIdeMode: vi.fn(() => false),
-      getAcpMode: vi.fn(() => true),
+      getExperimentalZedIntegration: vi.fn(() => true),
       getScreenReader: vi.fn(() => false),
       getGeminiMdFileCount: vi.fn(() => 0),
       getProjectRoot: vi.fn(() => '/'),

packages/cli/src/integration-tests/modelSteering.test.tsx

@@ -65,6 +65,10 @@ describe('Model Steering Integration', () => {
     // Resolve list_directory (Proceed)
     await rig.resolveTool('ReadFolder');
 
+    // Wait for the model to process the hint and output the next action
+    // Based on steering.responses, it should first acknowledge the hint
+    await rig.waitForOutput('ACK: I will focus on .txt files now.');
+
     // Then it should proceed with the next action
     await rig.waitForOutput(
       /Since you want me to focus on .txt files,[\s\S]*I will read file1.txt/,

packages/cli/src/services/BuiltinCommandLoader.test.ts

@@ -73,17 +73,7 @@ vi.mock('../ui/commands/agentsCommand.js', () => ({
 }));
 vi.mock('../ui/commands/bugCommand.js', () => ({ bugCommand: {} }));
 vi.mock('../ui/commands/chatCommand.js', () => ({
-  chatCommand: {
-    name: 'chat',
-    subCommands: [
-      { name: 'list' },
-      { name: 'save' },
-      { name: 'resume' },
-      { name: 'delete' },
-      { name: 'share' },
-      { name: 'checkpoints', hidden: true, subCommands: [{ name: 'list' }] },
-    ],
-  },
+  chatCommand: { name: 'chat', subCommands: [] },
   debugCommand: { name: 'debug' },
 }));
 vi.mock('../ui/commands/clearCommand.js', () => ({ clearCommand: {} }));
@@ -104,19 +94,7 @@ vi.mock('../ui/commands/modelCommand.js', () => ({
 }));
 vi.mock('../ui/commands/privacyCommand.js', () => ({ privacyCommand: {} }));
 vi.mock('../ui/commands/quitCommand.js', () => ({ quitCommand: {} }));
-vi.mock('../ui/commands/resumeCommand.js', () => ({
-  resumeCommand: {
-    name: 'resume',
-    subCommands: [
-      { name: 'list' },
-      { name: 'save' },
-      { name: 'resume' },
-      { name: 'delete' },
-      { name: 'share' },
-      { name: 'checkpoints', hidden: true, subCommands: [{ name: 'list' }] },
-    ],
-  },
-}));
+vi.mock('../ui/commands/resumeCommand.js', () => ({ resumeCommand: {} }));
 vi.mock('../ui/commands/statsCommand.js', () => ({ statsCommand: {} }));
 vi.mock('../ui/commands/themeCommand.js', () => ({ themeCommand: {} }));
 vi.mock('../ui/commands/toolsCommand.js', () => ({ toolsCommand: {} }));
@@ -278,7 +256,7 @@ describe('BuiltinCommandLoader', () => {
   });
 
   describe('chat debug command', () => {
-    it('should NOT add debug subcommand to chat/resume commands if not a nightly build', async () => {
+    it('should NOT add debug subcommand to chatCommand if not a nightly build', async () => {
       vi.mocked(isNightly).mockResolvedValue(false);
       const loader = new BuiltinCommandLoader(mockConfig);
       const commands = await loader.loadCommands(new AbortController().signal);
@@ -287,30 +265,9 @@ describe('BuiltinCommandLoader', () => {
       expect(chatCmd?.subCommands).toBeDefined();
       const hasDebug = chatCmd!.subCommands!.some((c) => c.name === 'debug');
       expect(hasDebug).toBe(false);
-
-      const resumeCmd = commands.find((c) => c.name === 'resume');
-      const resumeHasDebug =
-        resumeCmd?.subCommands?.some((c) => c.name === 'debug') ?? false;
-      expect(resumeHasDebug).toBe(false);
-
-      const chatCheckpointsCmd = chatCmd?.subCommands?.find(
-        (c) => c.name === 'checkpoints',
-      );
-      const chatCheckpointHasDebug =
-        chatCheckpointsCmd?.subCommands?.some((c) => c.name === 'debug') ??
-        false;
-      expect(chatCheckpointHasDebug).toBe(false);
-
-      const resumeCheckpointsCmd = resumeCmd?.subCommands?.find(
-        (c) => c.name === 'checkpoints',
-      );
-      const resumeCheckpointHasDebug =
-        resumeCheckpointsCmd?.subCommands?.some((c) => c.name === 'debug') ??
-        false;
-      expect(resumeCheckpointHasDebug).toBe(false);
     });
 
-    it('should add debug subcommand to chat/resume commands if it is a nightly build', async () => {
+    it('should add debug subcommand to chatCommand if it is a nightly build', async () => {
       vi.mocked(isNightly).mockResolvedValue(true);
       const loader = new BuiltinCommandLoader(mockConfig);
       const commands = await loader.loadCommands(new AbortController().signal);
@@ -319,27 +276,6 @@ describe('BuiltinCommandLoader', () => {
       expect(chatCmd?.subCommands).toBeDefined();
       const hasDebug = chatCmd!.subCommands!.some((c) => c.name === 'debug');
       expect(hasDebug).toBe(true);
-
-      const resumeCmd = commands.find((c) => c.name === 'resume');
-      const resumeHasDebug =
-        resumeCmd?.subCommands?.some((c) => c.name === 'debug') ?? false;
-      expect(resumeHasDebug).toBe(true);
-
-      const chatCheckpointsCmd = chatCmd?.subCommands?.find(
-        (c) => c.name === 'checkpoints',
-      );
-      const chatCheckpointHasDebug =
-        chatCheckpointsCmd?.subCommands?.some((c) => c.name === 'debug') ??
-        false;
-      expect(chatCheckpointHasDebug).toBe(true);
-
-      const resumeCheckpointsCmd = resumeCmd?.subCommands?.find(
-        (c) => c.name === 'checkpoints',
-      );
-      const resumeCheckpointHasDebug =
-        resumeCheckpointsCmd?.subCommands?.some((c) => c.name === 'debug') ??
-        false;
-      expect(resumeCheckpointHasDebug).toBe(true);
     });
   });
 });

packages/cli/src/services/BuiltinCommandLoader.ts

@@ -31,7 +31,6 @@ import { docsCommand } from '../ui/commands/docsCommand.js';
 import { directoryCommand } from '../ui/commands/directoryCommand.js';
 import { editorCommand } from '../ui/commands/editorCommand.js';
 import { extensionsCommand } from '../ui/commands/extensionsCommand.js';
-import { footerCommand } from '../ui/commands/footerCommand.js';
 import { helpCommand } from '../ui/commands/helpCommand.js';
 import { shortcutsCommand } from '../ui/commands/shortcutsCommand.js';
 import { rewindCommand } from '../ui/commands/rewindCommand.js';
@@ -78,41 +77,6 @@ export class BuiltinCommandLoader implements ICommandLoader {
     const handle = startupProfiler.start('load_builtin_commands');
 
     const isNightlyBuild = await isNightly(process.cwd());
-    const addDebugToChatResumeSubCommands = (
-      subCommands: SlashCommand[] | undefined,
-    ): SlashCommand[] | undefined => {
-      if (!subCommands) {
-        return subCommands;
-      }
-
-      const withNestedCompatibility = subCommands.map((subCommand) => {
-        if (subCommand.name !== 'checkpoints') {
-          return subCommand;
-        }
-
-        return {
-          ...subCommand,
-          subCommands: addDebugToChatResumeSubCommands(subCommand.subCommands),
-        };
-      });
-
-      if (!isNightlyBuild) {
-        return withNestedCompatibility;
-      }
-
-      return withNestedCompatibility.some(
-        (cmd) => cmd.name === debugCommand.name,
-      )
-        ? withNestedCompatibility
-        : [
-            ...withNestedCompatibility,
-            { ...debugCommand, suggestionGroup: 'checkpoints' },
-          ];
-    };
-
-    const chatResumeSubCommands = addDebugToChatResumeSubCommands(
-      chatCommand.subCommands,
-    );
 
     const allDefinitions: Array<SlashCommand | null> = [
       aboutCommand,
@@ -121,7 +85,9 @@ export class BuiltinCommandLoader implements ICommandLoader {
       bugCommand,
       {
         ...chatCommand,
-        subCommands: chatResumeSubCommands,
+        subCommands: isNightlyBuild
+          ? [...(chatCommand.subCommands || []), debugCommand]
+          : chatCommand.subCommands,
       },
       clearCommand,
       commandsCommand,
@@ -153,7 +119,6 @@ export class BuiltinCommandLoader implements ICommandLoader {
           ]
         : [extensionsCommand(this.config?.getEnableExtensionReloading())]),
       helpCommand,
-      footerCommand,
       shortcutsCommand,
       ...(this.config?.getEnableHooksUI() ? [hooksCommand] : []),
       rewindCommand,
@@ -188,10 +153,7 @@ export class BuiltinCommandLoader implements ICommandLoader {
       ...(isDevelopment ? [profileCommand] : []),
       quitCommand,
       restoreCommand(this.config),
-      {
-        ...resumeCommand,
-        subCommands: addDebugToChatResumeSubCommands(resumeCommand.subCommands),
-      },
+      resumeCommand,
       statsCommand,
       themeCommand,
       toolsCommand,

packages/cli/src/services/CommandService.test.ts

@@ -17,9 +17,21 @@ const createMockCommand = (name: string, kind: CommandKind): SlashCommand => ({
   action: vi.fn(),
 });
 
+const mockCommandA = createMockCommand('command-a', CommandKind.BUILT_IN);
+const mockCommandB = createMockCommand('command-b', CommandKind.BUILT_IN);
+const mockCommandC = createMockCommand('command-c', CommandKind.FILE);
+const mockCommandB_Override = createMockCommand('command-b', CommandKind.FILE);
+
 class MockCommandLoader implements ICommandLoader {
-  constructor(private readonly commands: SlashCommand[]) {}
-  loadCommands = vi.fn(async () => Promise.resolve(this.commands));
+  private commandsToLoad: SlashCommand[];
+
+  constructor(commandsToLoad: SlashCommand[]) {
+    this.commandsToLoad = commandsToLoad;
+  }
+
+  loadCommands = vi.fn(
+    async (): Promise<SlashCommand[]> => Promise.resolve(this.commandsToLoad),
+  );
 }
 
 describe('CommandService', () => {
@@ -31,74 +43,424 @@ describe('CommandService', () => {
     vi.restoreAllMocks();
   });
 
-  describe('basic loading', () => {
-    it('should aggregate commands from multiple successful loaders', async () => {
-      const cmdA = createMockCommand('a', CommandKind.BUILT_IN);
-      const cmdB = createMockCommand('b', CommandKind.USER_FILE);
-      const service = await CommandService.create(
-        [new MockCommandLoader([cmdA]), new MockCommandLoader([cmdB])],
-        new AbortController().signal,
-      );
+  it('should load commands from a single loader', async () => {
+    const mockLoader = new MockCommandLoader([mockCommandA, mockCommandB]);
+    const service = await CommandService.create(
+      [mockLoader],
+      new AbortController().signal,
+    );
 
-      expect(service.getCommands()).toHaveLength(2);
-      expect(service.getCommands()).toEqual(
-        expect.arrayContaining([cmdA, cmdB]),
-      );
-    });
+    const commands = service.getCommands();
 
-    it('should handle empty loaders and failed loaders gracefully', async () => {
-      const cmdA = createMockCommand('a', CommandKind.BUILT_IN);
-      const failingLoader = new MockCommandLoader([]);
-      vi.spyOn(failingLoader, 'loadCommands').mockRejectedValue(
-        new Error('fail'),
-      );
+    expect(mockLoader.loadCommands).toHaveBeenCalledTimes(1);
+    expect(commands).toHaveLength(2);
+    expect(commands).toEqual(
+      expect.arrayContaining([mockCommandA, mockCommandB]),
+    );
+  });
 
-      const service = await CommandService.create(
-        [
-          new MockCommandLoader([cmdA]),
-          new MockCommandLoader([]),
-          failingLoader,
-        ],
-        new AbortController().signal,
-      );
+  it('should aggregate commands from multiple loaders', async () => {
+    const loader1 = new MockCommandLoader([mockCommandA]);
+    const loader2 = new MockCommandLoader([mockCommandC]);
+    const service = await CommandService.create(
+      [loader1, loader2],
+      new AbortController().signal,
+    );
 
-      expect(service.getCommands()).toHaveLength(1);
-      expect(service.getCommands()[0].name).toBe('a');
-      expect(debugLogger.debug).toHaveBeenCalledWith(
-        'A command loader failed:',
-        expect.any(Error),
-      );
-    });
+    const commands = service.getCommands();
 
-    it('should return a readonly array of commands', async () => {
-      const service = await CommandService.create(
-        [new MockCommandLoader([createMockCommand('a', CommandKind.BUILT_IN)])],
-        new AbortController().signal,
-      );
-      expect(() => (service.getCommands() as unknown[]).push({})).toThrow();
-    });
+    expect(loader1.loadCommands).toHaveBeenCalledTimes(1);
+    expect(loader2.loadCommands).toHaveBeenCalledTimes(1);
+    expect(commands).toHaveLength(2);
+    expect(commands).toEqual(
+      expect.arrayContaining([mockCommandA, mockCommandC]),
+    );
+  });
 
-    it('should pass the abort signal to all loaders', async () => {
-      const controller = new AbortController();
-      const loader = new MockCommandLoader([]);
-      await CommandService.create([loader], controller.signal);
-      expect(loader.loadCommands).toHaveBeenCalledWith(controller.signal);
+  it('should override commands from earlier loaders with those from later loaders', async () => {
+    const loader1 = new MockCommandLoader([mockCommandA, mockCommandB]);
+    const loader2 = new MockCommandLoader([
+      mockCommandB_Override,
+      mockCommandC,
+    ]);
+    const service = await CommandService.create(
+      [loader1, loader2],
+      new AbortController().signal,
+    );
+
+    const commands = service.getCommands();
+
+    expect(commands).toHaveLength(3); // Should be A, C, and the overridden B.
+
+    // The final list should contain the override from the *last* loader.
+    const commandB = commands.find((cmd) => cmd.name === 'command-b');
+    expect(commandB).toBeDefined();
+    expect(commandB?.kind).toBe(CommandKind.FILE); // Verify it's the overridden version.
+    expect(commandB).toEqual(mockCommandB_Override);
+
+    // Ensure the other commands are still present.
+    expect(commands).toEqual(
+      expect.arrayContaining([
+        mockCommandA,
+        mockCommandC,
+        mockCommandB_Override,
+      ]),
+    );
+  });
+
+  it('should handle loaders that return an empty array of commands gracefully', async () => {
+    const loader1 = new MockCommandLoader([mockCommandA]);
+    const emptyLoader = new MockCommandLoader([]);
+    const loader3 = new MockCommandLoader([mockCommandB]);
+    const service = await CommandService.create(
+      [loader1, emptyLoader, loader3],
+      new AbortController().signal,
+    );
+
+    const commands = service.getCommands();
+
+    expect(emptyLoader.loadCommands).toHaveBeenCalledTimes(1);
+    expect(commands).toHaveLength(2);
+    expect(commands).toEqual(
+      expect.arrayContaining([mockCommandA, mockCommandB]),
+    );
+  });
+
+  it('should load commands from successful loaders even if one fails', async () => {
+    const successfulLoader = new MockCommandLoader([mockCommandA]);
+    const failingLoader = new MockCommandLoader([]);
+    const error = new Error('Loader failed');
+    vi.spyOn(failingLoader, 'loadCommands').mockRejectedValue(error);
+
+    const service = await CommandService.create(
+      [successfulLoader, failingLoader],
+      new AbortController().signal,
+    );
+
+    const commands = service.getCommands();
+    expect(commands).toHaveLength(1);
+    expect(commands).toEqual([mockCommandA]);
+    expect(debugLogger.debug).toHaveBeenCalledWith(
+      'A command loader failed:',
+      error,
+    );
+  });
+
+  it('getCommands should return a readonly array that cannot be mutated', async () => {
+    const service = await CommandService.create(
+      [new MockCommandLoader([mockCommandA])],
+      new AbortController().signal,
+    );
+
+    const commands = service.getCommands();
+
+    // Expect it to throw a TypeError at runtime because the array is frozen.
+    expect(() => {
+      // @ts-expect-error - Testing immutability is intentional here.
+      commands.push(mockCommandB);
+    }).toThrow();
+
+    // Verify the original array was not mutated.
+    expect(service.getCommands()).toHaveLength(1);
+  });
+
+  it('should pass the abort signal to all loaders', async () => {
+    const controller = new AbortController();
+    const signal = controller.signal;
+
+    const loader1 = new MockCommandLoader([mockCommandA]);
+    const loader2 = new MockCommandLoader([mockCommandB]);
+
+    await CommandService.create([loader1, loader2], signal);
+
+    expect(loader1.loadCommands).toHaveBeenCalledTimes(1);
+    expect(loader1.loadCommands).toHaveBeenCalledWith(signal);
+    expect(loader2.loadCommands).toHaveBeenCalledTimes(1);
+    expect(loader2.loadCommands).toHaveBeenCalledWith(signal);
+  });
+
+  it('should rename extension commands when they conflict', async () => {
+    const builtinCommand = createMockCommand('deploy', CommandKind.BUILT_IN);
+    const userCommand = createMockCommand('sync', CommandKind.FILE);
+    const extensionCommand1 = {
+      ...createMockCommand('deploy', CommandKind.FILE),
+      extensionName: 'firebase',
+      description: '[firebase] Deploy to Firebase',
+    };
+    const extensionCommand2 = {
+      ...createMockCommand('sync', CommandKind.FILE),
+      extensionName: 'git-helper',
+      description: '[git-helper] Sync with remote',
+    };
+
+    const mockLoader1 = new MockCommandLoader([builtinCommand]);
+    const mockLoader2 = new MockCommandLoader([
+      userCommand,
+      extensionCommand1,
+      extensionCommand2,
+    ]);
+
+    const service = await CommandService.create(
+      [mockLoader1, mockLoader2],
+      new AbortController().signal,
+    );
+
+    const commands = service.getCommands();
+    expect(commands).toHaveLength(4);
+
+    // Built-in command keeps original name
+    const deployBuiltin = commands.find(
+      (cmd) => cmd.name === 'deploy' && !cmd.extensionName,
+    );
+    expect(deployBuiltin).toBeDefined();
+    expect(deployBuiltin?.kind).toBe(CommandKind.BUILT_IN);
+
+    // Extension command conflicting with built-in gets renamed
+    const deployExtension = commands.find(
+      (cmd) => cmd.name === 'firebase.deploy',
+    );
+    expect(deployExtension).toBeDefined();
+    expect(deployExtension?.extensionName).toBe('firebase');
+
+    // User command keeps original name
+    const syncUser = commands.find(
+      (cmd) => cmd.name === 'sync' && !cmd.extensionName,
+    );
+    expect(syncUser).toBeDefined();
+    expect(syncUser?.kind).toBe(CommandKind.FILE);
+
+    // Extension command conflicting with user command gets renamed
+    const syncExtension = commands.find(
+      (cmd) => cmd.name === 'git-helper.sync',
+    );
+    expect(syncExtension).toBeDefined();
+    expect(syncExtension?.extensionName).toBe('git-helper');
+  });
+
+  it('should handle user/project command override correctly', async () => {
+    const builtinCommand = createMockCommand('help', CommandKind.BUILT_IN);
+    const userCommand = createMockCommand('help', CommandKind.FILE);
+    const projectCommand = createMockCommand('deploy', CommandKind.FILE);
+    const userDeployCommand = createMockCommand('deploy', CommandKind.FILE);
+
+    const mockLoader1 = new MockCommandLoader([builtinCommand]);
+    const mockLoader2 = new MockCommandLoader([
+      userCommand,
+      userDeployCommand,
+      projectCommand,
+    ]);
+
+    const service = await CommandService.create(
+      [mockLoader1, mockLoader2],
+      new AbortController().signal,
+    );
+
+    const commands = service.getCommands();
+    expect(commands).toHaveLength(2);
+
+    // User command overrides built-in
+    const helpCommand = commands.find((cmd) => cmd.name === 'help');
+    expect(helpCommand).toBeDefined();
+    expect(helpCommand?.kind).toBe(CommandKind.FILE);
+
+    // Project command overrides user command (last wins)
+    const deployCommand = commands.find((cmd) => cmd.name === 'deploy');
+    expect(deployCommand).toBeDefined();
+    expect(deployCommand?.kind).toBe(CommandKind.FILE);
+  });
+
+  it('should handle secondary conflicts when renaming extension commands', async () => {
+    // User has both /deploy and /gcp.deploy commands
+    const userCommand1 = createMockCommand('deploy', CommandKind.FILE);
+    const userCommand2 = createMockCommand('gcp.deploy', CommandKind.FILE);
+
+    // Extension also has a deploy command that will conflict with user's /deploy
+    const extensionCommand = {
+      ...createMockCommand('deploy', CommandKind.FILE),
+      extensionName: 'gcp',
+      description: '[gcp] Deploy to Google Cloud',
+    };
+
+    const mockLoader = new MockCommandLoader([
+      userCommand1,
+      userCommand2,
+      extensionCommand,
+    ]);
+
+    const service = await CommandService.create(
+      [mockLoader],
+      new AbortController().signal,
+    );
+
+    const commands = service.getCommands();
+    expect(commands).toHaveLength(3);
+
+    // Original user command keeps its name
+    const deployUser = commands.find(
+      (cmd) => cmd.name === 'deploy' && !cmd.extensionName,
+    );
+    expect(deployUser).toBeDefined();
+
+    // User's dot notation command keeps its name
+    const gcpDeployUser = commands.find(
+      (cmd) => cmd.name === 'gcp.deploy' && !cmd.extensionName,
+    );
+    expect(gcpDeployUser).toBeDefined();
+
+    // Extension command gets renamed with suffix due to secondary conflict
+    const deployExtension = commands.find(
+      (cmd) => cmd.name === 'gcp.deploy1' && cmd.extensionName === 'gcp',
+    );
+    expect(deployExtension).toBeDefined();
+    expect(deployExtension?.description).toBe('[gcp] Deploy to Google Cloud');
+  });
+
+  it('should handle multiple secondary conflicts with incrementing suffixes', async () => {
+    // User has /deploy, /gcp.deploy, and /gcp.deploy1
+    const userCommand1 = createMockCommand('deploy', CommandKind.FILE);
+    const userCommand2 = createMockCommand('gcp.deploy', CommandKind.FILE);
+    const userCommand3 = createMockCommand('gcp.deploy1', CommandKind.FILE);
+
+    // Extension has a deploy command
+    const extensionCommand = {
+      ...createMockCommand('deploy', CommandKind.FILE),
+      extensionName: 'gcp',
+      description: '[gcp] Deploy to Google Cloud',
+    };
+
+    const mockLoader = new MockCommandLoader([
+      userCommand1,
+      userCommand2,
+      userCommand3,
+      extensionCommand,
+    ]);
+
+    const service = await CommandService.create(
+      [mockLoader],
+      new AbortController().signal,
+    );
+
+    const commands = service.getCommands();
+    expect(commands).toHaveLength(4);
+
+    // Extension command gets renamed with suffix 2 due to multiple conflicts
+    const deployExtension = commands.find(
+      (cmd) => cmd.name === 'gcp.deploy2' && cmd.extensionName === 'gcp',
+    );
+    expect(deployExtension).toBeDefined();
+    expect(deployExtension?.description).toBe('[gcp] Deploy to Google Cloud');
+  });
+
+  it('should report conflicts via getConflicts', async () => {
+    const builtinCommand = createMockCommand('deploy', CommandKind.BUILT_IN);
+    const extensionCommand = {
+      ...createMockCommand('deploy', CommandKind.FILE),
+      extensionName: 'firebase',
+    };
+
+    const mockLoader = new MockCommandLoader([
+      builtinCommand,
+      extensionCommand,
+    ]);
+
+    const service = await CommandService.create(
+      [mockLoader],
+      new AbortController().signal,
+    );
+
+    const conflicts = service.getConflicts();
+    expect(conflicts).toHaveLength(1);
+
+    expect(conflicts[0]).toMatchObject({
+      name: 'deploy',
+      winner: builtinCommand,
+      losers: [
+        {
+          renamedTo: 'firebase.deploy',
+          command: expect.objectContaining({
+            name: 'deploy',
+            extensionName: 'firebase',
+          }),
+        },
+      ],
     });
   });
 
-  describe('conflict delegation', () => {
-    it('should delegate conflict resolution to SlashCommandResolver', async () => {
-      const builtin = createMockCommand('help', CommandKind.BUILT_IN);
-      const user = createMockCommand('help', CommandKind.USER_FILE);
+  it('should report extension vs extension conflicts correctly', async () => {
+    // Both extensions try to register 'deploy'
+    const extension1Command = {
+      ...createMockCommand('deploy', CommandKind.FILE),
+      extensionName: 'firebase',
+    };
+    const extension2Command = {
+      ...createMockCommand('deploy', CommandKind.FILE),
+      extensionName: 'aws',
+    };
 
-      const service = await CommandService.create(
-        [new MockCommandLoader([builtin, user])],
-        new AbortController().signal,
-      );
+    const mockLoader = new MockCommandLoader([
+      extension1Command,
+      extension2Command,
+    ]);
 
-      expect(service.getCommands().map((c) => c.name)).toContain('help');
-      expect(service.getCommands().map((c) => c.name)).toContain('user.help');
-      expect(service.getConflicts()).toHaveLength(1);
+    const service = await CommandService.create(
+      [mockLoader],
+      new AbortController().signal,
+    );
+
+    const conflicts = service.getConflicts();
+    expect(conflicts).toHaveLength(1);
+
+    expect(conflicts[0]).toMatchObject({
+      name: 'deploy',
+      winner: expect.objectContaining({
+        name: 'deploy',
+        extensionName: 'firebase',
+      }),
+      losers: [
+        {
+          renamedTo: 'aws.deploy', // ext2 is 'aws' and it lost because it was second in the list
+          command: expect.objectContaining({
+            name: 'deploy',
+            extensionName: 'aws',
+          }),
+        },
+      ],
     });
   });
+
+  it('should report multiple conflicts for the same command name', async () => {
+    const builtinCommand = createMockCommand('deploy', CommandKind.BUILT_IN);
+    const ext1 = {
+      ...createMockCommand('deploy', CommandKind.FILE),
+      extensionName: 'ext1',
+    };
+    const ext2 = {
+      ...createMockCommand('deploy', CommandKind.FILE),
+      extensionName: 'ext2',
+    };
+
+    const mockLoader = new MockCommandLoader([builtinCommand, ext1, ext2]);
+
+    const service = await CommandService.create(
+      [mockLoader],
+      new AbortController().signal,
+    );
+
+    const conflicts = service.getConflicts();
+    expect(conflicts).toHaveLength(1);
+    expect(conflicts[0].name).toBe('deploy');
+    expect(conflicts[0].losers).toHaveLength(2);
+    expect(conflicts[0].losers).toEqual(
+      expect.arrayContaining([
+        expect.objectContaining({
+          renamedTo: 'ext1.deploy',
+          command: expect.objectContaining({ extensionName: 'ext1' }),
+        }),
+        expect.objectContaining({
+          renamedTo: 'ext2.deploy',
+          command: expect.objectContaining({ extensionName: 'ext2' }),
+        }),
+      ]),
+    );
+  });
 });